{"slug": "kimi-k3-1-pro-highspeed-on-amazon-bedrock-model-ids-caching-flex-tier-and", "title": "Kimi K3.1 Pro Highspeed on Amazon Bedrock: Model IDs, Caching, Flex Tier, and OpenCode", "summary": "Amazon Web Services made Moonshot AI's Kimi K3.1 Pro Highspeed available on Amazon Bedrock on November 10, 2026, with the recommended model ID `global.moonshotai.kimi-k3.1` and residency profiles `us.moonshotai.kimi-k3.1` and `in.moonshotai.kimi-k3.1`. AWS said the Highspeed serving variant delivers two to three times the output speed of Kimi K3.1 Pro on Moonshot's side, supports a 1M-token context, always-on reasoning with `reasoning_effort` set to `low`, `high` or `max` (default `max`), and is reachable through OpenAI-compatible Chat Completions, Responses, Bedrock Converse and Invoke. AWS stated that inference data stays within the AWS data boundary, is not shared with Moonshot AI and is not used to train the model, with a half-price Flex tier, explicit prompt caching and an official OpenCode `opencode.json` example offered as cost and tooling levers.", "body_md": "**Amazon Web Services put [Kimi K3.1 Pro Highspeed](https://aws.amazon.com/blogs/machine-learning/introducing-kimi-k3-1-pro-on-amazon-bedrock/?ref=explainx) from Moonshot AI on Amazon Bedrock on November 10, 2026.** You reach it through cross-Region inference profiles, the same OpenAI-compatible Chat Completions path AWS already documents for Kimi K3, plus Converse, Responses, explicit prompt caching, a half-price Flex tier, and a first-party OpenCode wiring guide.\n\nFor teams that already run agents on Bedrock, the practical change is a model ID swap and a chance to test Moonshot’s post-trained K3.1 line without sending traffic to Moonshot’s own API. For teams that evaluated Kimi only on the [Kimi API](https://explainx.ai/blog/kimi-k3-1-pro-open-weights-benchmarks-price-terminal-bench-nov-4-2026), Bedrock is the procurement path: IAM, regional profiles, and AWS’s statement that inference data stays inside the AWS boundary and is not used to train the model.\n\n## TL;DR\n\n| Item | Detail | \n|---|---|\n| Announced | November 10, 2026 ( [AWS ML blog](https://aws.amazon.com/blogs/machine-learning/introducing-kimi-k3-1-pro-on-amazon-bedrock/?ref=explainx) ) | \n| Variant on Bedrock | Kimi K3.1 Pro **Highspeed** (faster output token rate) | \n| Recommended model ID | `global.moonshotai.kimi-k3.1` | \n| Residency profiles | `us.moonshotai.kimi-k3.1` ,`in.moonshotai.kimi-k3.1` | \n| Context | 1M tokens (same class as K3 per AWS/Moonshot) | \n| APIs | OpenAI Chat Completions (recommended), Responses, Bedrock Converse and Invoke | \n| Reasoning | Always on; `reasoning_effort` or`reasoning.effort` :`low` ,`high` ,`max` (default`max` ) | \n| Cost levers | Flex tier (half Standard price), explicit prompt caching | \n| Agent IDE | Official OpenCode `opencode.json` example in the AWS post | \n\n## What Kimi K3.1 Pro Highspeed is\n\nMoonshot released **Kimi K3.1 Pro** in October 2026 as the first model in its K3.1 series. explainx.ai covered the launch benchmarks, API price, and the November 4 open-weight drop in [the Kimi K3.1 Pro overview](https://explainx.ai/blog/kimi-k3-1-pro-open-weights-benchmarks-price-terminal-bench-nov-4-2026) and the earlier [Kimi K3 open-weights explainer](https://explainx.ai/blog/kimi-k3-open-weights-2-8-trillion-parameters-july-2026).\n\nAWS’s November post positions **Highspeed** as the Bedrock serving SKU: same endpoint family as Kimi K3 on Bedrock, but Moonshot reports **two to three times** the output speed of Kimi K3.1 Pro on Moonshot’s side. That matters for coding agents, which chain many short generations. Each tool-call loop waits on output tokens; a faster stream shortens edit-test-fix cycles.\n\nMoonshot’s own blog claims gains in long-horizon agentic work, fewer factual errors, financial modeling, office documents, and coding for game engines, 3D scenes, and CAD scripts. Those are vendor statements. Independent scores such as the [Artificial Analysis line in our launch post](https://explainx.ai/blog/kimi-k3-1-pro-open-weights-benchmarks-price-terminal-bench-nov-4-2026) are useful, but Bedrock adoption is really a hosting and governance decision.\n\n## Why Bedrock instead of the Kimi API\n\nThree reasons enterprises cite when a frontier open-weight model lands on a hyperscaler:\n\n1. **Contract and IAM.** The model rides an existing AWS agreement. Security reviews that already approved Bedrock do not restart for Moonshot’s corporate entity.\n2. **Data boundary language.** AWS states that your data is processed within the AWS data boundary, is**not shared with Moonshot AI** , and is**not used to train** the model. It also lists zero data retention for inference and zero operator access during inference. Read the[Bedrock service terms](https://aws.amazon.com/bedrock/security-compliance/?ref=explainx) for your account type; this post does not restate legal guarantees.\n3. **Model mixing.** Bedrock already hosts other coding-oriented models. Our[GLM 5.3 on Bedrock guide](https://explainx.ai/blog/glm-5-3-amazon-bedrock-enterprise-coding-model-ids-2026) shows the same OpenAI-client pattern. You can A/B Kimi K3.1 against GLM or Claude family models behind one billing relationship.\n\nMoonshot still operates the [Kimi API at roughly $2/$10 per million tokens](https://explainx.ai/blog/kimi-k3-1-pro-open-weights-benchmarks-price-terminal-bench-nov-4-2026) for direct access and open-weight self-hosting. Bedrock pricing is on [Amazon Bedrock pricing](https://aws.amazon.com/bedrock/pricing/?ref=explainx) and may differ by tier and cache hits. Compare **total cost per successful agent task**, not list price alone.\n\nFor agent security context on Kimi-class models, see [Vals AI’s cyber-offense bench](https://explainx.ai/blog/vals-ai-cyberoffense-gpt-6-sol-daybreak-deepseek-v4-1-flash-open-weight-cyber-range-2026), which scored Kimi K3.1 Pro in a controlled range, and Moonshot’s own cyber ablation in the launch post.\n\n## Model IDs and regions\n\nAWS exposes Kimi K3.1 Pro Highspeed only through **cross-Region inference profiles**:\n\n| Profile ID | Routing | When to use | \n|---|---|---|\n| `global.moonshotai.kimi-k3.1` | Any supported commercial AWS Region | Default; **lower per-token cost** per AWS | \n| `us.moonshotai.kimi-k3.1` | US Regions only | US data residency | \n| `in.moonshotai.kimi-k3.1` | India Regions only | India residency | \n\nIf you already call Kimi K3 on Bedrock, AWS says you can **change the model ID** in existing code. Regional availability and card details are on AWS’s model documentation linked from the announcement.\n\n## Calling Kimi with the OpenAI SDK\n\nAWS recommends the **OpenAI-compatible Chat Completions** endpoint on `bedrock-runtime`. You authenticate with a short-lived bearer token from AWS credentials via `aws-bedrock-token-generator`, not a long-lived OpenAI key.\n\n``` python\nfrom openai import OpenAI\nfrom aws_bedrock_token_generator import provide_token\n\nREGION = \"us-west-2\"\nMODEL_ID = \"global.moonshotai.kimi-k3.1\"\n\nclient = OpenAI(\n    base_url=f\"https://bedrock-runtime.{REGION}.amazonaws.com/openai/v1\",\n    api_key=provide_token(region=REGION),\n)\n\ncompletion = client.chat.completions.create(\n    model=MODEL_ID,\n    messages=[\n        {\"role\": \"system\", \"content\": \"You are a concise senior engineer.\"},\n        {\"role\": \"user\", \"content\": \"When should I prefer a queue over a stream?\"},\n    ],\n    reasoning_effort=\"high\",\n)\nprint(completion.choices[0].message.content)\n```\n\n**Reasoning effort:** Kimi K3.1 Pro Highspeed always reasons before answering. `reasoning_effort` accepts `low`, `high`, or `max`. If you omit it, the model uses **` max`**, which increases latency and output tokens. For routing inside an agent, use `low` on classification steps and `high` or `max` on codegen.\n\nThe **Responses API** uses the same client with `reasoning={\"effort\": \"low\"}` instead. The **Converse API** uses boto3 with `modelId=\"global.moonshotai.kimi-k3.1\"` and may return a reasoning block before text; pick the text block from the content array as AWS’s sample shows.\n\nPrerequisites from AWS: Python 3.11+, `openai`, `aws-bedrock-token-generator`, `boto3`, and IAM permission for `bedrock:InvokeModel` and streaming, plus `bedrock:CallWithBearerToken` when using bearer tokens.\n\n## Flex tier: half price, slower tail\n\nBedrock’s **Flex** tier serves Kimi at **half the Standard on-demand price** with potentially longer response times. AWS suggests batch evals, summarization, and background agent steps.\n\nOn Chat Completions, pass `service_tier=\"flex\"`. On Converse, set `serviceTier` to `{\"type\": \"flex\"}`. Standard, Priority, and Flex draw from the **same on-demand quota**. AWS recommends starting with a small Flex share and watching CloudWatch dimensions `ServiceTier` and `ResolvedServiceTier` before shifting traffic.\n\nThis mirrors the tier story on [GLM 5.3 Bedrock](https://explainx.ai/blog/glm-5-3-amazon-bedrock-enterprise-coding-model-ids-2026) and the broader [Bedrock tokenomics patterns](https://explainx.ai/blog/amazon-bedrock-tokenomics-data-lake-finops-2026) we covered for finops teams.\n\n## Explicit prompt caching for agents\n\nAgents resend large stable prefixes: system prompts, tool schemas, repo guidelines, and reference docs. Kimi K3.1 Pro Highspeed supports **automatic** prefix caching and **explicit** breakpoints on Chat Completions, Responses, and Invoke.\n\nRules from AWS:\n\n- Explicit caching **does not** work on**Converse** .\n- A `prompt_cache_breakpoint` must follow**at least 256 tokens** of stable content.\n- Cached content is retained **at least 30 minutes** .\n- Cache **writes** bill higher than standard input;**reads** bill at a discount and**do not** count toward input TPM quota.\n\nThe AWS post shows a Responses example that marks breakpoints on a long `repo_guidelines.md` system prompt and on the user turn, then prints `cached_tokens` from `usage.input_tokens_details`. That pattern is the same class of savings we described for [prompt caching on Bedrock with GPT-6 Astra patterns](https://explainx.ai/blog/aws-bedrock-prompt-caching-openai-gpt-6-astra-agent-patterns-62-percent-input-cost-2026).\n\nFor multi-turn tool use on the open weights, Moonshot’s model card notes you must pass back the full assistant message including `reasoning_content`. Bedrock callers should confirm the same requirement in their harness when migrating from the Kimi API.\n\n## OpenCode on Bedrock\n\nAWS documents a terminal coding agent path with **OpenCode** and the Bedrock Chat Completions provider. You point `baseURL` at `https://bedrock-runtime.us-west-2.amazonaws.com/openai/v1` (swap Region), register `global.moonshotai.kimi-k3.1` with reasoning capabilities, and set it as the default model. Short-lived **Bedrock API keys** from the console are recommended for production; long-term keys are for exploration.\n\n## OpenCode connect and reasoning variants\n\nOpenCode connect dialog for Kimi K3.1 Pro Highspeed on Amazon Bedrock Chat Completions\n*Image: [Amazon Web Services](https://aws.amazon.com/blogs/machine-learning/introducing-kimi-k3-1-pro-on-amazon-bedrock/?ref=explainx)*\n\nAfter `/connect` and key paste, OpenCode stores credentials once per key. Use `/models` to switch between Kimi and other providers mid-session.\n\nReasoning effort variants for Kimi K3.1 Pro Highspeed in OpenCode on Bedrock\n*Image: [Amazon Web Services](https://aws.amazon.com/blogs/machine-learning/introducing-kimi-k3-1-pro-on-amazon-bedrock/?ref=explainx)*\n\nThe `capabilities` block lists supported reasoning efforts so OpenCode exposes **low**, **high**, and **max** variants. That UI matters when you want a fast refactor pass on `low` and a careful migration on `max` without editing config between tasks.\n\nOpenCode is one of several harnesses in our [agent harness comparisons](https://explainx.ai/blog/how-to-run-glm-5-2-coding-plan-agent-harnesses-2026). Bedrock is not the only backend: Moonshot’s own Kimi Code product and third-party hosts remain options.\n\n## Comparison snapshot (where to look elsewhere)\n\n| Need | Consider | \n|---|---|\n| Lowest friction if you already pay Moonshot | Kimi API direct ( [pricing page](https://platform.kimi.ai/docs/pricing/chat?ref=explainx) ) | \n| AWS-only procurement and residency profiles | **Bedrock Kimi K3.1 Highspeed** (this announcement) | \n| Self-host with open weights | [Hugging Face moonshotai/Kimi-K3.1-Pro](https://explainx.ai/blog/kimi-k3-1-pro-open-weights-benchmarks-price-terminal-bench-nov-4-2026) plus vLLM/SGLang | \n| Another open-weight coder on the same bill | [GLM 5.3 on Bedrock](https://explainx.ai/blog/glm-5-3-amazon-bedrock-enterprise-coding-model-ids-2026) | \n| Cyber eval context | [Vals cyber-offense bench](https://explainx.ai/blog/vals-ai-cyberoffense-gpt-6-sol-daybreak-deepseek-v4-1-flash-open-weight-cyber-range-2026) | \n\nAWS does not publish a side-by-side Bedrock bench table in this post. Run your own tasks on your repositories before switching production traffic.\n\n## What to verify in your account\n\n1. **Entitlement.** New Bedrock models sometimes roll out in waves. Confirm Kimi K3.1 appears in the console Playground and in`ListFoundationModels` for your Region.\n2. **Reasoning defaults.** Leaving effort at default`max` can balloon cost on simple steps. Set effort per call type in your router.\n3. **Cache economics.** Measure cache hit rate before assuming savings; cache writes cost more upfront.\n4. **License if you later self-host.** Moonshot’s open license includes revenue gates for large Model-as-a-Service operators, separate from Bedrock terms ([details in our weights update](https://explainx.ai/blog/kimi-k3-1-pro-open-weights-benchmarks-price-terminal-bench-nov-4-2026) ).\n5. **Speed claims.** The 2–3× speed statement is Moonshot’s comparison to Kimi K3.1 Pro on Moonshot infrastructure, not a guarantee against other Bedrock models.\n\n## Bedrock console quick start\n\nIf you are not ready to wire Python yet, AWS’s path is: open the **Amazon Bedrock** console, choose **Test**, then **Playground**, and select **Kimi K3.1** from the model picker. That is the fastest way to sanity-check entitlement and reasoning defaults before you paste IAM policies into CI. When the playground works, copy the same model ID into your agent router and enable streaming so tool-calling harnesses see tokens as they arrive. Teams that standardize on [Amazon Bedrock AgentCore evaluation patterns](https://explainx.ai/blog/amazon-bedrock-agentcore-evaluation-playbook-deepeval-autoeval-guide-2026) can add Kimi as another model arm in the same eval harness once the ID is approved.", "url": "https://wpnews.pro/news/kimi-k3-1-pro-highspeed-on-amazon-bedrock-model-ids-caching-flex-tier-and", "canonical_source": "https://explainx.ai/blog/kimi-k3-1-pro-amazon-bedrock-highspeed-opencode-guide-2026?utm_source=rss&utm_medium=feed&utm_campaign=blog", "published_at": "2026-11-11 00:00:00+00:00", "updated_at": "2026-11-11 03:24:11.360197+00:00", "lang": "en", "topics": ["large-language-models", "ai-products", "ai-agents", "ai-infrastructure", "ai-tools"], "entities": ["Amazon Web Services", "Amazon Bedrock", "Moonshot AI", "Kimi K3.1 Pro Highspeed", "Kimi K3", "OpenCode", "AWS Converse API"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/kimi-k3-1-pro-highspeed-on-amazon-bedrock-model-ids-caching-flex-tier-and", "markdown": "https://wpnews.pro/news/kimi-k3-1-pro-highspeed-on-amazon-bedrock-model-ids-caching-flex-tier-and.md", "text": "https://wpnews.pro/news/kimi-k3-1-pro-highspeed-on-amazon-bedrock-model-ids-caching-flex-tier-and.txt", "jsonld": "https://wpnews.pro/news/kimi-k3-1-pro-highspeed-on-amazon-bedrock-model-ids-caching-flex-tier-and.jsonld"}}