{"slug": "llama-3-1-70b-api-access", "title": "Llama 3.1 70B API access?", "summary": "Hugging Face users are experiencing confusion over Llama 3.1 70B API access via Inference Providers like Featherless. The issue is likely a provider-specific model availability mismatch rather than a general access problem, as access to the gated repo does not guarantee the provider serves the exact model ID and task. Users should verify the exact model ID (e.g., Instruct variant) and check provider support before debugging tokens or curl syntax.", "body_md": "There’s been a lot of confusion around how Inference Providers are supposed to be used:\n\nI don’t think “You have been granted access to this model” necessarily contradicts “Model not supported by provider featherless-ai”.\n\nThe short version is:\n\n| Check | What it means |\n|---|---|\n| “You have been granted access” on the model page | You have access to the gated model repo / weights / model page resources. |\n| The browser widget works | Some provider/path available from the widget could run something for that page. It does not necessarily prove your third-party app is using the same provider, model id, task, token scope, or endpoint. |\n`Model not supported by provider featherless-ai` |\nThe selected provider, here `featherless-ai` , may not currently expose the exact model id and task that your API call is asking for. |\n\nSo I would first check the **exact model id + provider + task** combination before debugging the token or curl syntax too much.\n\nThe quickest first check is the model search page with the Inference Providers filter:\n\n[https://huggingface.co/models?inference_provider=all](https://huggingface.co/models?inference_provider=all)\n\nThen search for the exact model id and, if needed, narrow the provider filter to Featherless. If the exact model/provider combination is not listed there, changing the curl call probably will not make that provider serve the model.\n\nAlso, one subtle point: `meta-llama/Llama-3.1-70B`\n\nand `meta-llama/Llama-3.1-70B-Instruct`\n\nare not interchangeable.\n\n`meta-llama/Llama-3.1-70B`\n\n`meta-llama/Llama-3.1-70B-Instruct`\n\nIf your third-party app is making chat-completion-style calls, I would first verify whether the **Instruct** variant is available through the provider you are trying to use, rather than assuming that access to the base repo means the provider can serve it through chat completions.\n\nA practical order of checks would be:\n\nConfirm the exact model id:\n\n`meta-llama/Llama-3.1-70B`\n\n`meta-llama/Llama-3.1-70B-Instruct`\n\n`meta-llama/Meta-Llama-3.1-70B-Instruct`\n\nCheck whether that exact model is currently exposed through Inference Providers:\n\nIf you are explicitly forcing Featherless, try not forcing it:\n\n`provider=\"auto\"`\n\nin `huggingface_hub`\n\n, or`:featherless-ai`\n\nsuffix if you are using the OpenAI-compatible router model name.If it works with `auto`\n\nbut fails with `featherless-ai`\n\n, that suggests a provider-specific availability/mapping issue, not a general Llama access issue.\n\nCheck the local client version if you are using Python:\n\n``` python\npython -c \"import huggingface_hub; print(huggingface_hub.__version__)\"\n```\n\nFeatherless’ HF integration post says to use `huggingface_hub`\n\nv0.33.0 or newer:\n\nIf you still get the error, the useful info to post back would be:\n\n`provider=\"featherless-ai\"`\n\n, `:featherless-ai`\n\n, or `auto`\n\n`huggingface_hub`\n\nversion, if applicableTo summarize my guess: this is probably not “you do not have access to Llama” in the simple gated-repo sense. It is more likely one of these:\n\n`featherless-ai`\n\n,The first thing I would rule out is the non-fixable case: **is the exact model id currently available through the provider you are forcing?**", "url": "https://wpnews.pro/news/llama-3-1-70b-api-access", "canonical_source": "https://discuss.huggingface.co/t/llama-3-1-70b-api-access/177106#post_2", "published_at": "2026-06-23 22:22:51+00:00", "updated_at": "2026-06-24 00:41:40.202219+00:00", "lang": "en", "topics": ["large-language-models", "ai-tools", "developer-tools"], "entities": ["Hugging Face", "Featherless", "Meta", "Llama 3.1 70B", "Llama 3.1 70B Instruct"], "alternates": {"html": "https://wpnews.pro/news/llama-3-1-70b-api-access", "markdown": "https://wpnews.pro/news/llama-3-1-70b-api-access.md", "text": "https://wpnews.pro/news/llama-3-1-70b-api-access.txt", "jsonld": "https://wpnews.pro/news/llama-3-1-70b-api-access.jsonld"}}