# Model APIs and inference for AI agents

> 8 model APIs & inference listings ranked by the Anchor benchmark. Leader OpenAI API (A). Frontier and open-weight model APIs, fast inference hardware and routers that sit in front of many providers. Compared on price, context, tool use, data handling and how often models get retired under you.

- Canonical: https://www.anchorterminal.com/categories/inference
- Markdown: https://www.anchorterminal.com/categories/inference.md (~5,350 tokens)
- Slim: https://www.anchorterminal.com/categories/inference.min.md (~430 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/categories/inference.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- Updated: 2026-10-05

Frontier and open-weight model APIs, fast inference hardware and routers that sit in front of many providers. Compared on price, context, tool use, data handling and how often models get retired under you.

- Tools ranked: 8 · agent-ready (BB or better): 5 · accept x402: 1 · hosted endpoints: 7 · desk reviews by the panel: 26
- JSON: https://www.anchorterminal.com/api/v1/tools.json (list) · https://www.anchorterminal.com/api/v1/rankings.json (ranked) · https://www.anchorterminal.com/api/v1/x402.json (payable) · https://www.anchorterminal.com/api/v1/capabilities.json (by capability)
- Grades run AA, A, BB, B, C, D, E, F · methodology: https://www.anchorterminal.com/benchmark/

- Capabilities in this category: inference.llm, inference.router, inference.fast, inference.open-weights
- https://letme.dev/inference.llm picks the top-graded tool in this list and says how to call it direct; calling through letme comes later (https://www.anchorterminal.com/letme/index.md)

## Ranking

| # | Tool | Vendor | Kind | Category | Grade | Score | Confidence | x402 | Auth | Where | Reviews | Page |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 2 | OpenAI API | OpenAI | Model API | Models | A | 82.8 | medium | no | API key | hosted | 3.5/5 (8) | https://www.anchorterminal.com/tools/openai-api.md |
| 18 | Claude API | Anthropic | Model API | Models | BB | 77.6 | medium | no | API key | hosted | none | https://www.anchorterminal.com/tools/anthropic-api.md |
| 31 | GroqCloud | Groq | Model API | Models | BB | 75.7 | medium | no | API key | hosted | 3.5/5 (8) | https://www.anchorterminal.com/tools/groq.md |
| 68 | BlockRun.AI | BlockRun, Inc. | MCP server | Models | BB | 72.5 | medium | yes | OAuth or key | local | 3.5/5 (2) | https://www.anchorterminal.com/tools/blockrun-ai.md |
| 86 | Mistral AI API | Mistral AI | Model API | Models | BB | 71.3 | medium | no | API key | hosted | 4/5 (2) | https://www.anchorterminal.com/tools/mistral-api.md |
| 119 | OpenRouter | OpenRouter | Model router | Models | B | 68.8 | medium | no | API key | hosted | 3/5 (2) | https://www.anchorterminal.com/tools/openrouter.md |
| 223 | Gemini Developer API | Google | Model API | Models | B | 62 | medium | no | API key | hosted | 3.5/5 (2) | https://www.anchorterminal.com/tools/gemini-api.md |
| 386 | DeepSeek API | DeepSeek | Model API | Models | D | 47.1 | medium | no | API key | hosted | 2/5 (2) | https://www.anchorterminal.com/tools/deepseek-api.md |

Scores are from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/), with Performance and Task success pending. p95 latency and context cost come from our probes, which haven't run yet.

## Summaries

### 2. OpenAI API, A (82.8)

OpenAI's API for accessing its models through Responses, Chat Completions and Batch endpoints. Official OpenAPI document and an llms.txt index. Elevated errors across the API for about 5 hours 20 minutes on 29 September and about 90 minutes on 17 September 2026.

- Page: https://www.anchorterminal.com/tools/openai-api · Markdown: https://www.anchorterminal.com/tools/openai-api.md · JSON: https://www.anchorterminal.com/api/v1/tools/openai-api.json
- Capabilities: inference.llm · endpoint: `https://api.openai.com/v1`

### 18. Claude API, BB (77.6)

Anthropic's Messages API for Claude, with server-side tools, an MCP connector and computer use. Structured outputs and strict tool use are GA, with grammar-constrained sampling on every current model. Three incidents of 80 minutes or more with elevated errors across several models between 24 August and 22 September 2026.

- Page: https://www.anchorterminal.com/tools/anthropic-api · Markdown: https://www.anchorterminal.com/tools/anthropic-api.md · JSON: https://www.anchorterminal.com/api/v1/tools/anthropic-api.json
- Capabilities: inference.llm · endpoint: `https://api.anthropic.com/v1/messages`

### 31. GroqCloud, BB (75.7)

OpenAI-compatible inference API serving open-weight models on Groq's processors. Free plan with no card, at 30 requests a minute and 1,000 a day on gpt-oss. Four model shutdown dates between 2026-07-17 and 2026-09-21, with no stated minimum notice.

- Page: https://www.anchorterminal.com/tools/groq · Markdown: https://www.anchorterminal.com/tools/groq.md · JSON: https://www.anchorterminal.com/api/v1/tools/groq.json
- Capabilities: inference.llm, inference.fast, inference.open-weights · endpoint: `https://api.groq.com/openai/v1`

### 68. BlockRun.AI, BB (72.5)

Model and API gateway with an OpenAI-compatible interface, MCP support and per-request stablecoin payments through x402. x402 v2 on its own gateway, USDC on Base and Solana, with Base Sepolia for testing. The status page shows live checks only and keeps no incident history.

- Page: https://www.anchorterminal.com/tools/blockrun-ai · Markdown: https://www.anchorterminal.com/tools/blockrun-ai.md · JSON: https://www.anchorterminal.com/api/v1/tools/blockrun-ai.json
- Capabilities: inference.llm, web.search, market.crypto

### 86. Mistral AI API, BB (71.3)

Mistral's API for its open-weight and proprietary models, with EU and US regional endpoints. Public OpenAPI document at docs.mistral.ai/openapi.yaml and an llms.txt with Markdown twins. Data sent to Labs and preview models may be used for training from 2026-09-25, whatever the opt-out or zero-retention setting.

- Page: https://www.anchorterminal.com/tools/mistral-api · Markdown: https://www.anchorterminal.com/tools/mistral-api.md · JSON: https://www.anchorterminal.com/api/v1/tools/mistral-api.json
- Capabilities: inference.llm, inference.open-weights · endpoint: `https://api.mistral.ai/v1`

### 119. OpenRouter, B (68.8)

One OpenAI-compatible API in front of 80+ model providers, at the providers' own token prices, with fallbacks and routing controls. Public OpenAPI document, llms.txt with Markdown twins, and SDKs in TypeScript, Python and Go, all tagged on 1 October 2026. No machine payment. USDC top-ups go to a prepaid balance.

- Page: https://www.anchorterminal.com/tools/openrouter · Markdown: https://www.anchorterminal.com/tools/openrouter.md · JSON: https://www.anchorterminal.com/api/v1/tools/openrouter.json
- Capabilities: inference.llm, inference.router · endpoint: `https://openrouter.ai/api/v1`

### 223. Gemini Developer API, B (62)

Google's API for Gemini models, including content generation and agent interactions. Free tier on 3.8 Flash and Flash-Lite with no billing account. Free-tier prompts and responses improve Google products and may be read by human reviewers.

- Page: https://www.anchorterminal.com/tools/gemini-api · Markdown: https://www.anchorterminal.com/tools/gemini-api.md · JSON: https://www.anchorterminal.com/api/v1/tools/gemini-api.json
- Capabilities: inference.llm · endpoint: `https://generativelanguage.googleapis.com/v1beta`

### 386. DeepSeek API, D (47.1)

Hangzhou lab's API for the V4 model family, in both OpenAI and Anthropic request formats, with off-peak pricing. Accepts both OpenAI and Anthropic request formats. 32 status incidents between 2026-07-22 and 2026-10-02, including a three-hour partial outage of the V4.1 Flash API on 2026-09-14.

- Page: https://www.anchorterminal.com/tools/deepseek-api · Markdown: https://www.anchorterminal.com/tools/deepseek-api.md · JSON: https://www.anchorterminal.com/api/v1/tools/deepseek-api.json
- Capabilities: inference.llm, inference.open-weights · endpoint: `https://api.deepseek.com`

## Indexed, not reviewed (287)

Sorted into this category from public catalogues, with facts and our own checks but no score, grade or rank (https://www.anchorterminal.com/indexed/index.md).

| Listing | Kind | What it does | Why it's here |
| --- | --- | --- | --- |
| [3DPACK.ING — Container & Truck Load Planning](https://www.anchorterminal.com/tools/3dpack-container-loading.md) | MCP server | Container load planning in Claude & ChatGPT: what fits, how full, a labelled 3D plan in the chat. | vendor's own |
| [3dstreet](https://www.anchorterminal.com/tools/3dstreet.md) | MCP server | Drive an open 3DStreet scene tab from Claude Desktop or Claude Code via MCP tool calls. | vendor's own |
| [AdButler](https://www.anchorterminal.com/tools/adbutler-mcp-server.md) | MCP server | Manage AdButler campaigns, zones, ad items, VAST, and reports from Claude, ChatGPT, and Cursor. | vendor's own |
| [adtest-mcp](https://www.anchorterminal.com/tools/adtest-mcp.md) | MCP server | AI-only ad scoring for Claude: score image, video or text ads on 13 dimensions before you spend. | vendor's own |
| [agentcentral](https://www.anchorterminal.com/tools/agentcentral-amazon-seller-central.md) | MCP server | Hosted Amazon Seller Central and Amazon Ads MCP server for Claude, ChatGPT, Cursor, and agents. | vendor's own |
| [Aion](https://www.anchorterminal.com/tools/aion.md) | model family | Aion 3.5 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses… | on OpenRouter |
| [Aion Mini](https://www.anchorterminal.com/tools/aion-mini.md) | model family | Aion 3.5 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It… | on OpenRouter |
| [Aion-2](https://www.anchorterminal.com/tools/aion-2.md) | model family | Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling. It is particularly strong… | on OpenRouter |
| [Aion-3](https://www.anchorterminal.com/tools/aion-3.md) | model family | Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses… | on OpenRouter |
| [Aion-3.0-Mini](https://www.anchorterminal.com/tools/aion-3-0-mini.md) | model family | Aion-3.0 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the DeepSeek family of… | on OpenRouter |
| [Aion-RP](https://www.anchorterminal.com/tools/aion-rp.md) | model family | Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a… | on OpenRouter |
| [ai·rete·rag](https://www.anchorterminal.com/tools/ai-rete-rag-mcp.md) | MCP server | Author rules from policy docs, then decide: a Rete engine gives the verdict, an LLM explains why. | vendor's own |
| [augments.dev MCP server](https://www.anchorterminal.com/tools/augments-mcp.md) | MCP server | Augments MCP Server - A comprehensive framework documentation provider for Claude Code | vendor's own |
| [AXF](https://www.anchorterminal.com/tools/smartergpt-axf.md) | MCP server | Agent eXoskeleton Framework control plane for workspace-native agent capabilities over MCP. | vendor's own |
| [Axis](https://www.anchorterminal.com/tools/useaxis-axis.md) | MCP server | Coding agents from Claude Code, Cursor and Codex claim jobs and lock files on one shared board. | vendor's own |
| [baton](https://www.anchorterminal.com/tools/kasabeh-baton.md) | MCP server | Pass the baton between coding agents: convert sessions across Claude Code, OpenCode, Codex & more. | vendor's own |
| [Call Me](https://www.anchorterminal.com/tools/callmemcp-call-me.md) | MCP server | Your AI rings your iPhone, speaks its question, and gets your spoken answer back as text. | vendor's own |
| [Claude Fable](https://www.anchorterminal.com/tools/claude-fable.md) | model family | Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running… | on OpenRouter |
| [Claude FAF](https://www.anchorterminal.com/tools/claude-faf-mcp.md) | MCP server | Persistent project context for Claude. IANA-registered .faf format. | vendor's own |
| [Claude Haiku](https://www.anchorterminal.com/tools/claude-haiku.md) | model family | Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction… | on OpenRouter |
| [Claude Opus](https://www.anchorterminal.com/tools/claude-opus.md) | model family | Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work,… | on OpenRouter |
| [Claude Sonnet](https://www.anchorterminal.com/tools/claude-sonnet.md) | model family | Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a… | on OpenRouter |
| [Cleanor](https://www.anchorterminal.com/tools/cleanor.md) | MCP server | Zero-auth MCP: image optimize, cited storage/format data, and dev utilities LLMs get wrong. | vendor's own |
| [Code Pathfinder](https://www.anchorterminal.com/tools/codepathfinder-pathfinder.md) | MCP server | Code intelligence MCP server: call graphs, type inference, and symbol search for Python/Go. | vendor's own |
| [Codestral](https://www.anchorterminal.com/tools/codestral.md) | model family | Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency,… | on OpenRouter |
| [Command A](https://www.anchorterminal.com/tools/command-a.md) | model family | Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance… | on OpenRouter |
| [Command A+](https://www.anchorterminal.com/tools/command-a-2.md) | model family | Command A+ is Cohere's flagship model for enterprise agentic workflows. It accepts text and image inputs with a 192K… | on OpenRouter |
| [Command R](https://www.anchorterminal.com/tools/command-r.md) | model family | command-r-08-2024 is an update of the \[Command R\](/models/cohere/command-r) with improved performance for multilingual… | on OpenRouter |
| [Command R+](https://www.anchorterminal.com/tools/command-r-2.md) | model family | command-r-plus-08-2024 is an update of the \[Command R+\](/models/cohere/command-r-plus) with roughly 50% higher… | on OpenRouter |
| [Command R7B](https://www.anchorterminal.com/tools/command-r7b.md) | model family | Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG,… | on OpenRouter |
| [crisphive.com MCP server](https://www.anchorterminal.com/tools/crisphive-mcp.md) | MCP server | Field operations on a deterministic solver — run jobs, crews & fleet from Claude or ChatGPT. | vendor's own |
| [Cydonia V4](https://www.anchorterminal.com/tools/cydonia-v4.md) | model family | Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and… | on OpenRouter |
| [DataDoe MCP](https://www.anchorterminal.com/tools/datadoe-amazon-seller-mcp.md) | MCP server | Hosted Amazon Seller and Vendor MCP server for Claude, ChatGPT, Cursor, Codex, Gemini, Copilot. | vendor's own |
| [DeepSeek V3](https://www.anchorterminal.com/tools/deepseek-v3.md) | model family | DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and… | on OpenRouter |
| [DeepSeek V3 Terminus](https://www.anchorterminal.com/tools/deepseek-v3-terminus.md) | model family | DeepSeek-V3.1 Terminus is an update to \[DeepSeek V3.1\](/deepseek/deepseek-chat-v3.1) that maintains the model's… | on OpenRouter |
| [DeepSeek V4 Flash](https://www.anchorterminal.com/tools/deepseek-v4-flash.md) | model family | DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal… | on OpenRouter |
| [DeepSeek V4 Flash Vision](https://www.anchorterminal.com/tools/deepseek-v4-flash-vision.md) | model family | DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of \[DeepSeek V4 Flash… | on OpenRouter |
| [DeepSeek V4 Pro](https://www.anchorterminal.com/tools/deepseek-v4-pro.md) | model family | DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro. | on OpenRouter |
| [Devstral](https://www.anchorterminal.com/tools/devstral.md) | model family | Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter… | on OpenRouter |
| [Dropl MCP](https://www.anchorterminal.com/tools/dropl-mcp.md) | MCP server | Create client-editable photo and video galleries from Cursor or Claude Code, then embed them. | vendor's own |
| [Ember-1](https://www.anchorterminal.com/tools/ember-1.md) | model family | Ember-1 is a specialized reasoning model from Fireworks Research, built on \[Kimi… | on OpenRouter |
| [ERNIE VL](https://www.anchorterminal.com/tools/ernie-vl.md) | model family | ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B… | on OpenRouter |
| [etincel-nonfiction](https://www.anchorterminal.com/tools/etincel-nonfiction.md) | MCP server | Trainable non-fiction writing voice, presets, and an anti-AI-tells audit for Claude, via MCP. | vendor's own |
| [FAQai](https://www.anchorterminal.com/tools/faqai-mcp-server.md) | MCP server | Turn any document into a production-ready RAG dataset from Claude, Cursor, or any MCP host. | vendor's own |
| [flowcontent.io MCP server](https://www.anchorterminal.com/tools/flowcontent-mcp-server.md) | MCP server | Connect Claude, Cursor, or any MCP client to your FlowContent account | vendor's own |
| [Fugu Max](https://www.anchorterminal.com/tools/fugu-max.md) | model family | Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a… | on OpenRouter |
| [Fugu Ultra](https://www.anchorterminal.com/tools/fugu-ultra.md) | model family | Fugu Ultra is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a… | on OpenRouter |
| [Fugu Ultra v2](https://www.anchorterminal.com/tools/fugu-ultra-v2.md) | model family | Fugu Ultra v2 is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu… | on OpenRouter |
| [gammainfra.com MCP server](https://www.anchorterminal.com/tools/gammainfra-mcp-server.md) | MCP server | Smart LLM routing across every major provider via one OpenAI-shape API. | vendor's own |
| [Garmin — MissingMCP](https://www.anchorterminal.com/tools/missingmcp-garmin.md) | MCP server | Garmin data in Claude: 135 tools — activities, sleep, HRV, training, workouts. Free, open source. | vendor's own |
| [gate](https://www.anchorterminal.com/tools/spideriq-gate.md) | MCP server | SpiderIQ Gate: SpiderGate LLM gateway (completions, models, usage, traces) | vendor's own |
| [Gemini FAF](https://www.anchorterminal.com/tools/gemini-faf-mcp.md) | MCP server | Persistent project context for Google Gemini. Python/FastMCP. IANA-registered .faf format. | vendor's own |
| [Gemini Flash](https://www.anchorterminal.com/tools/gemini-flash.md) | model family | Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software… | on OpenRouter |
| [Gemini Flash Lite](https://www.anchorterminal.com/tools/gemini-flash-lite.md) | model family | Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for… | on OpenRouter |
| [Gemini Pro](https://www.anchorterminal.com/tools/gemini-pro.md) | model family | Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance,… | on OpenRouter |
| [Gemini Pro 06-05](https://www.anchorterminal.com/tools/gemini-pro-06-05.md) | model family | Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and… | on OpenRouter |
| [Gemini Pro Custom Tools](https://www.anchorterminal.com/tools/gemini-pro-custom-tools.md) | model family | Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing… | on OpenRouter |
| [Gemma](https://www.anchorterminal.com/tools/gemma.md) | model family | Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total… | on OpenRouter |
| [Glean Remote MCP Server](https://www.anchorterminal.com/tools/glean-mcp.md) | MCP server | Remote MCP Server that securely connects Glean Enterprise Knowledge with your IDE, LLM, or agents. | vendor's own |
| [GLM](https://www.anchorterminal.com/tools/z-ai-glm.md) | model family | GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent… | on OpenRouter |

The first 60 of 287, most used first; all of them are in https://www.anchorterminal.com/categories/inference.json and https://www.anchorterminal.com/api/v1/indexed.json (`category`).