{"slug": "show-hn-modelparams-dev-ai-model-parameters-database-ui-api-npm", "title": "Show HN: Modelparams.dev: AI Model Parameters Database (UI, API, NPM)", "summary": "Modelparams.dev launched an open, community-maintained catalog of LLM model parameters, offering a UI, API, and NPM package. The database currently covers 191 models from providers like OpenAI, detailing parameters such as max tokens, temperature, and top P.", "body_md": "#\nEvery LLM parameter,\n\nfor every model.\n\nAn open, community-maintained catalog of LLM model parameters. Browse the UI below, query the API, or install the npm package.\n\nAccess type\n\nProviders\n\nParameters\n\n191 of 191 models\n\n## OpenAI\n\n##\n[Chatgpt 4o Latest](/models/openai/chatgpt-4o-latest)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 3.5 Turbo](/models/openai/gpt-3.5-turbo)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 4 Turbo](/models/openai/gpt-4-turbo)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 4 Turbo 2024-04-09](/models/openai/gpt-4-turbo-2024-04-09)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 4.1](/models/openai/gpt-4.1)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 4.1 Mini](/models/openai/gpt-4.1-mini)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 4.1 Nano](/models/openai/gpt-4.1-nano)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[GPT-4o](/models/openai/gpt-4o)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 4o 2024-11-20](/models/openai/gpt-4o-2024-11-20)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[GPT-4o mini](/models/openai/gpt-4o-mini)\nOpenAI\n3 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…2 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. | — |\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. | — |\n\n##\n[Gpt 5](/models/openai/gpt-5)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (minimal | low | medium | high) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5 Chat Latest](/models/openai/gpt-5-chat-latest)\nOpenAI\n1 param\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\n##\n[Gpt 5 Mini](/models/openai/gpt-5-mini)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (minimal | low | medium | high) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5 Nano](/models/openai/gpt-5-nano)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (minimal | low | medium | high) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.1](/models/openai/gpt-5.1)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (none | low | medium | high) | \"none\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.1 Codex Max](/models/openai/gpt-5.1-codex-max-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.1 Codex](/models/openai/gpt-5.1-codex-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.2](/models/openai/gpt-5.2)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (none | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.2 Codex](/models/openai/gpt-5.2-codex-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.2](/models/openai/gpt-5.2-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.3 Codex](/models/openai/gpt-5.3-codex)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.3 Codex Spark](/models/openai/gpt-5.3-codex-spark-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.3 Codex](/models/openai/gpt-5.3-codex-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.4](/models/openai/gpt-5.4)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (none | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.4 Mini](/models/openai/gpt-5.4-mini)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (none | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.4 Mini](/models/openai/gpt-5.4-mini-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.4 Nano](/models/openai/gpt-5.4-nano)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (none | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.4 Pro](/models/openai/gpt-5.4-pro)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.4 Pro](/models/openai/gpt-5.4-pro-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.4](/models/openai/gpt-5.4-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.5](/models/openai/gpt-5.5)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (none | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.5 Pro](/models/openai/gpt-5.5-pro)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[Gpt 5.5 Pro](/models/openai/gpt-5.5-pro-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[Gpt 5.5](/models/openai/gpt-5.5-subscription)\nOpenAI\nSubscription\n3 params\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning.effort`\n|\nenum (minimal | low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n|\nReasoning summary\n`reasoning.summary`\n|\nenum (auto | concise | detailed | none) | \"auto\" | Controls the level of reasoning summary returned with the response. | — |\n\nOutput\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nVerbosity\n`text.verbosity`\n|\nenum (low | medium | high) | \"medium\" | Controls how concise or detailed the model's final text response should be. | — |\n\n##\n[o1](/models/openai/o1)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[o1-mini](/models/openai/o1-mini)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (minimal | low | medium | high) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[O1 Preview](/models/openai/o1-preview)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (minimal | low | medium | high) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[o3](/models/openai/o3)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[o3-mini](/models/openai/o3-mini)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[O3 Pro](/models/openai/o3-pro)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n##\n[o4-mini](/models/openai/o4-mini)\nOpenAI\n2 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax completion tokens\n`max_completion_tokens`\n|\ninteger (16…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nReasoning effort\n`reasoning_effort`\n|\nenum (low | medium | high | xhigh) | \"medium\" | Controls how much reasoning the model should perform before producing an answer. | — |\n\n## Anthropic\n\n##\n[Claude 3.5 Haiku 20241022](/models/anthropic/claude-3-5-haiku-20241022)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\n##\n[Claude 3.5 Haiku Latest](/models/anthropic/claude-3-5-haiku-latest)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\n##\n[Claude 3.5 Sonnet 20241022](/models/anthropic/claude-3-5-sonnet-20241022)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\n##\n[Claude 3.5 Sonnet Latest](/models/anthropic/claude-3-5-sonnet-latest)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\n##\n[Claude 3.7 Sonnet 20250219](/models/anthropic/claude-3-7-sonnet-20250219)\nAnthropic\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude 3.7 Sonnet Latest](/models/anthropic/claude-3-7-sonnet-latest)\nAnthropic\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude 3 Opus 20240229](/models/anthropic/claude-3-opus-20240229)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\n##\n[Claude 3 Opus Latest](/models/anthropic/claude-3-opus-latest)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\n##\n[Claude Fable 5](/models/anthropic/claude-fable-5)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (adaptive) | — | Only adaptive thinking is supported; omit the parameter entirely to run without thinking (an explicit disabled value is rejected). | — |\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"omitted\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"adaptive\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | xhigh | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Fable 5](/models/anthropic/claude-fable-5-subscription)\nAnthropic\nSubscription\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (adaptive) | — | Only adaptive thinking is supported; omit the parameter entirely to run without thinking (an explicit disabled value is rejected). | — |\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"omitted\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"adaptive\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | xhigh | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Haiku 4](/models/anthropic/claude-haiku-4)\nAnthropic\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Haiku 4.5](/models/anthropic/claude-haiku-4-5)\nAnthropic\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Haiku 4.5 20251001](/models/anthropic/claude-haiku-4-5-20251001)\nAnthropic\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\" or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\" or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Haiku 4.5 20251001](/models/anthropic/claude-haiku-4-5-20251001-subscription)\nAnthropic\nSubscription\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\" or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\" or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Haiku 4.5](/models/anthropic/claude-haiku-4-5-subscription)\nAnthropic\nSubscription\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Haiku 4](/models/anthropic/claude-haiku-4-subscription)\nAnthropic\nSubscription\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Opus 4.1 20250805](/models/anthropic/claude-opus-4-1-20250805)\nAnthropic\n7 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\" or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\" or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Opus 4.1 20250805](/models/anthropic/claude-opus-4-1-20250805-subscription)\nAnthropic\nSubscription\n7 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\" or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\" or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Opus 4 20250514](/models/anthropic/claude-opus-4-20250514)\nAnthropic\n7 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Opus 4 20250514](/models/anthropic/claude-opus-4-20250514-subscription)\nAnthropic\nSubscription\n7 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Opus 4.5 20251101](/models/anthropic/claude-opus-4-5-20251101)\nAnthropic\n8 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\" or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\" or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n4 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.5 20251101](/models/anthropic/claude-opus-4-5-20251101-subscription)\nAnthropic\nSubscription\n8 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\" or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\" or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n4 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.6](/models/anthropic/claude-opus-4-6)\nAnthropic\n8 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"enabled\", \"adaptive\"} or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"enabled\", \"adaptive\"} or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"enabled\", \"adaptive\"}\n|\n\nReasoning\n4 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.6](/models/anthropic/claude-opus-4-6-subscription)\nAnthropic\nSubscription\n8 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"enabled\", \"adaptive\"} or top_p ≠ null\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"enabled\", \"adaptive\"} or temperature ≠ null\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"enabled\", \"adaptive\"}\n|\n\nReasoning\n4 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.7](/models/anthropic/claude-opus-4-7)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"omitted\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"adaptive\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | xhigh | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.7](/models/anthropic/claude-opus-4-7-subscription)\nAnthropic\nSubscription\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"omitted\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"adaptive\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | xhigh | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.8](/models/anthropic/claude-opus-4-8)\nAnthropic\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"omitted\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"adaptive\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | xhigh | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4.8](/models/anthropic/claude-opus-4-8-subscription)\nAnthropic\nSubscription\n4 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"omitted\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"adaptive\"\n|\n|\nEffort\n`output_config.effort`\n|\nenum (low | medium | high | xhigh | max) | \"high\" | Controls Anthropic response thoroughness and token spend. | — |\n\n##\n[Claude Opus 4](/models/anthropic/claude-opus-4-subscription)\nAnthropic\nSubscription\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Sonnet 4 20250514](/models/anthropic/claude-sonnet-4-20250514)\nAnthropic\n7 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Sonnet 4 20250514](/models/anthropic/claude-sonnet-4-20250514-subscription)\nAnthropic\nSubscription\n7 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type = \"enabled\"\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type = \"enabled\"\n|\n\nReasoning\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. | — |\n|\nBudget tokens\n`thinking.budget_tokens`\n|\ninteger (1024…+∞) | 4096 | Maximum token budget Anthropic may use for extended thinking before producing the final answer. |\nOnly when thinking.type = \"enabled\"\n|\n|\nThinking display\n`thinking.display`\n|\nenum (summarized | omitted) | \"summarized\" | Controls whether Anthropic returns summarized or omitted thinking content. |\nOnly when thinking.type = \"enabled\"\n|\n\n##\n[Claude Sonnet 4.5](/models/anthropic/claude-sonnet-4-5)\nAnthropic\n6 params\n\nLength\n1 param\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nMax tokens\n`max_tokens`\n|\ninteger (1…+∞) | 4096 | Maximum number of output tokens the model may generate. | — |\n\nSampling\n3 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nTemperature\n`temperature`\n|\nnumber (0…1 step 0.1) | 1 | Controls randomness. Lower values make outputs more focused; higher values make them more varied. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n|\nTop P\n`top_p`\n|\nnumber (0…1 step 0.01) | 1 | Controls nucleus sampling by limiting generation to tokens whose cumulative probability reaches this value. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"} or temperature ≠ 1\n|\n|\nTop K\n`top_k`\n|\ninteger (0…+∞) | 0 | Limits token sampling to the top K most likely next tokens. |\nNot when thinking.type ∈ {\"adaptive\", \"enabled\"}\n|\n\nReasoning\n2 params\n\n| Parameter | Type | Default | Description | Condition |\n|---|---|---|---|---|\n|\nThinking mode\n`thinking.type`\n|\nenum (disabled | adaptive | enabled) | \"disabled\" | Controls the Anthropic thinking mode values supported by this model. |", "url": "https://wpnews.pro/news/show-hn-modelparams-dev-ai-model-parameters-database-ui-api-npm", "canonical_source": "https://modelparams.dev/", "published_at": "2026-06-18 15:07:03+00:00", "updated_at": "2026-06-18 15:22:45.400535+00:00", "lang": "en", "topics": ["large-language-models", "ai-tools", "developer-tools"], "entities": ["Modelparams.dev", "OpenAI", "ChatGPT 4o Latest", "GPT 3.5 Turbo", "GPT 4 Turbo", "GPT 4.1", "GPT 4.1 Mini"], "alternates": {"html": "https://wpnews.pro/news/show-hn-modelparams-dev-ai-model-parameters-database-ui-api-npm", "markdown": "https://wpnews.pro/news/show-hn-modelparams-dev-ai-model-parameters-database-ui-api-npm.md", "text": "https://wpnews.pro/news/show-hn-modelparams-dev-ai-model-parameters-database-ui-api-npm.txt", "jsonld": "https://wpnews.pro/news/show-hn-modelparams-dev-ai-model-parameters-database-ui-api-npm.jsonld"}}