{
  "listings": [
    {
      "slug": "anthropic-api",
      "name": "Claude API",
      "company": "Anthropic, PBC",
      "grade": "A",
      "fit": "yes",
      "why": "1M context is the default on current Opus, Sonnet, Fable and Haiku 5.5 with no beta header and standard pricing; Start tier allows 2,000,000 input tokens a minute on Opus and Sonnet (500,000 on Fable), so a 300K request fits.",
      "source": "https://platform.claude.com/docs/en/build-with-claude/context-windows",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/anthropic-api"
    },
    {
      "slug": "deepinfra",
      "name": "DeepInfra",
      "company": "Deep Infra Inc.",
      "grade": "B",
      "fit": "yes",
      "why": "DeepSeek V4.1 Flash is listed with 1M context, the only limit is 200 concurrent requests per model, and requests queue at capacity rather than 429.",
      "source": "https://docs.deepinfra.com/integrations/anthropic",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/deepinfra"
    },
    {
      "slug": "deepseek-api",
      "name": "DeepSeek API",
      "company": "Hangzhou DeepSeek Artificial Intelligence Co., Ltd.",
      "grade": "D",
      "fit": "yes",
      "why": "1M context with up to 384K output, limits are concurrency (2,500 and 500 per account) rather than tokens a minute, and off-peak pricing halves the cost of a large read.",
      "source": "https://api-docs.deepseek.com/quick_start/pricing",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/deepseek-api"
    },
    {
      "slug": "meta-model-api",
      "name": "Meta Model API",
      "company": "Meta Platforms, Inc.",
      "grade": "D",
      "fit": "yes",
      "why": "Muse Wonder and Muse Spark take 1,048,576 tokens with no long-context premium, and the per-model limit of 4,000,000 tokens a minute admits a 300K request.",
      "source": "https://dev.meta.ai/docs/models",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/meta-model-api"
    },
    {
      "slug": "amazon-bedrock",
      "name": "Amazon Bedrock",
      "company": "Amazon Web Services, Inc.",
      "grade": "A",
      "fit": "caveat",
      "why": "Whether a 300K request passes depends on the per-model tokens-per-minute quota (input plus output) and the account's daily cross-model cap in Service Quotas; increases may be denied unless existing quota is being consumed.",
      "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/quotas-runtime.html",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/amazon-bedrock"
    },
    {
      "slug": "azure-openai",
      "name": "Azure OpenAI in Foundry Models",
      "company": "Microsoft Corporation",
      "grade": "A",
      "fit": "caveat",
      "why": "Per-deployment tokens-per-minute quota must cover the request, Tier 0 subscriptions get no quota increases, and long-context requests are priced at 2x input on a pricing page that needs JavaScript.",
      "source": "https://learn.microsoft.com/en-us/azure/foundry/openai/quotas-limits",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/azure-openai"
    },
    {
      "slug": "blockrun-ai",
      "name": "BlockRun.AI",
      "company": "BlockRun, Inc.",
      "grade": "A",
      "fit": "caveat",
      "why": "No TPM or RPM cap on paid inference and 1M-context models are listed, but upstream calls time out at 120 seconds with a 504, which a 300K-token read can approach; max_tokens defaults to the smaller of model max and 8,192.",
      "source": "https://blockrun.ai/docs/api-reference/errors",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/blockrun-ai"
    },
    {
      "slug": "cloudflare-ai-gateway",
      "name": "Cloudflare AI Gateway",
      "company": "Cloudflare, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "Pass-through with no stated request token limit; the cacheable request size is 25 MB and the provider's own quota decides acceptance.",
      "source": "https://developers.cloudflare.com/ai-gateway/reference/limits/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cloudflare-ai-gateway"
    },
    {
      "slug": "eden-ai",
      "name": "Eden AI",
      "company": "Eden AI SAS",
      "grade": "C",
      "fit": "caveat",
      "why": "Acceptance depends on the upstream model; Eden states no token limit of its own beyond 10 requests a second.",
      "source": "https://www.edenai.co/docs/v3/overview/rate-limits",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/eden-ai"
    },
    {
      "slug": "fireworks-inference",
      "name": "Fireworks AI Inference",
      "company": "Fireworks.ai, Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "Kimi K3 lists 1,040K context, but per-model tokens-per-minute limits start at a default and grow 25 per cent every 15 minutes toward a tier ceiling, so a fresh account's first 300K request can be a 429.",
      "source": "https://docs.fireworks.ai/serverless/rate-limits",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/fireworks-inference"
    },
    {
      "slug": "gemini-api",
      "name": "Gemini Developer API",
      "company": "Google LLC",
      "grade": "B",
      "fit": "caveat",
      "why": "Gemini 4's 1M window is shared across input, thinking and output; acc-18 hit free-tier tokens-per-minute 429s past 300K and acc-14 found context preparation, not the call, was the bottleneck.",
      "source": "https://aistudio.google.com/docs/long-context",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/gemini-api"
    },
    {
      "slug": "hugging-face-inference-providers",
      "name": "Hugging Face Inference Providers",
      "company": "Hugging Face, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "The router lists context per provider (1M on some hosts, 65K on others for the same model), but router rate limits and 429 behaviour are undocumented, so acceptance of a 300K request is unknown.",
      "source": "https://router.huggingface.co/v1/models",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/hugging-face-inference-providers"
    },
    {
      "slug": "jais-3",
      "name": "Jais 3",
      "company": "Mohamed bin Zayed University of Artificial Intelligence (Institute of Foundation Models), with Inception42 (a G42 company)",
      "grade": "C",
      "fit": "caveat",
      "why": "512K context on 375B-A23B, but the reference deployment is 16 H100s or an 8-GPU H200 node, and high effort wants 32,768 output tokens on top.",
      "source": "https://huggingface.co/inception42/Jais-3-375B-A23B",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/jais-3"
    },
    {
      "slug": "litellm",
      "name": "LiteLLM",
      "company": "Berrie AI Incorporated",
      "grade": "B",
      "fit": "caveat",
      "why": "Pass-through; context_window_fallbacks reroute a request the primary rejects for length, otherwise the provider's quota decides.",
      "source": "https://docs.litellm.ai/docs/proxy/reliability",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/litellm"
    },
    {
      "slug": "minimax-api",
      "name": "MiniMax API",
      "company": "Nanonoble Pte. Ltd.",
      "grade": "C",
      "fit": "caveat",
      "why": "1M context and a 10,000,000 TPM limit admit a 300K request, but thinking defaults to max effort and counts against max_tokens (maximum 512K), so set effort and budget; latency is unverified.",
      "source": "https://platform.minimax.io/docs/guides/text-generation.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/minimax-api"
    },
    {
      "slug": "muse-spark-open",
      "name": "Muse Spark Open",
      "company": "Meta Platforms, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "1,048,576-token window, but the server's context length is fixed at launch, the recipes assume 4 to 8 B300 GPUs, and the sample commands set 32768.",
      "source": "https://dev.meta.ai/docs/muse-spark-open/settings.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/muse-spark-open"
    },
    {
      "slug": "nemotron-3-5-super-vl",
      "name": "Nemotron 3.5 Super VL",
      "company": "NVIDIA Corporation",
      "grade": "C",
      "fit": "caveat",
      "why": "The hosted page lists 262,144 tokens of context against the card's 1M and the hosted function caps completions at 131,072, so a 300K request fits only self-hosted.",
      "source": "https://build.nvidia.com/nvidia/nemotron-3.5-super-vl-120b-a12b.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/nemotron-3-5-super-vl"
    },
    {
      "slug": "openai-api",
      "name": "OpenAI API",
      "company": "OpenAI OpCo, LLC",
      "grade": "AA",
      "fit": "caveat",
      "why": "Reasoning tokens occupy the context window and the docs advise reserving 25,000 tokens; the pricing notes double input over 272K; acceptance of a 300K request depends on the tier's TPM shown in headers.",
      "source": "https://developers.openai.com/api/docs/guides/reasoning.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/openai-api"
    },
    {
      "slug": "openrouter",
      "name": "OpenRouter",
      "company": "OpenRouter, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "Depends on the routed model and host; max_tokens routes to hosts that fit it, and the in-flight credit budget can 429 a large request on a low-balance account.",
      "source": "https://openrouter.ai/docs/api_reference/limits.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/openrouter"
    },
    {
      "slug": "prism-inference",
      "name": "Prism Inference",
      "company": "Prism Technologies Inc",
      "grade": "C",
      "fit": "caveat",
      "why": "deepseek-v4.1-flash takes 1,000,000 tokens and oversize requests return context_length_exceeded rather than truncating; per-key limits aren't published, so a 300K call may hit 429.",
      "source": "https://prisminference.com/llms-full.txt",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/prism-inference"
    },
    {
      "slug": "ramp-router",
      "name": "Router.com (Ramp Router)",
      "company": "Ramp Business Corporation",
      "grade": "C",
      "fit": "caveat",
      "why": "Models expose context_window (1,050,000 on some) and a long_context_token_threshold of 272000 with higher rates; a fallback list is bounded by its smallest window and rate limits are unpublished.",
      "source": "https://docs.router.com/api/models.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/ramp-router"
    },
    {
      "slug": "reflection-beam",
      "name": "Beam 1.0",
      "company": "Reflection AI",
      "grade": "C",
      "fit": "caveat",
      "why": "Beam has a 1M context and 128K max output, but limits are per organisation in tokens per minute plus a daily token allowance renewing at 00:00 UTC, so a 300K call can spend the day's allowance.",
      "source": "https://developers.reflection.ai/models.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/reflection-beam"
    },
    {
      "slug": "routerplus",
      "name": "RouterPlus",
      "company": "EVO HQ Inc",
      "grade": "D",
      "fit": "caveat",
      "why": "context_overflow is a typed error that never fails over; trial keys run at 20 RPM and the balance hold assumes 32,768 output tokens, so send max_tokens with a large prompt.",
      "source": "https://app.routerplus.com/docs/limits.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/routerplus"
    },
    {
      "slug": "sambanova",
      "name": "SambaCloud",
      "company": "SambaNova Systems, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "Production models stop at 128K; only preview MiniMax-M3.1-Flash has 1M, and preview models have limited capacity and short-notice removal; Developer tier is capped at 20M tokens a day.",
      "source": "https://docs.sambanova.ai/docs/en/models/sambacloud-models.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/sambanova"
    },
    {
      "slug": "siliconflow",
      "name": "SiliconFlow",
      "company": "SiliconFlow Labs Pte. Ltd.",
      "grade": "E",
      "fit": "caveat",
      "why": "Context lengths are per model (the reasoning table shows 65,536 for GLM-4.1V-9B-Thinking) and output past max_tokens is cut with finish_reason length; the rate-limits page returns 404.",
      "source": "https://docs.siliconflow.com/en/userguide/capabilities/reasoning.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/siliconflow"
    },
    {
      "slug": "together-inference",
      "name": "Together AI Inference",
      "company": "Together Computer, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "Context depends on the model; serverless has no guaranteed capacity and bursts return 429 or 503, with provisioned throughput as the stated route.",
      "source": "https://docs.together.ai/docs/serverless/rate-limits.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/together-inference"
    },
    {
      "slug": "vercel-ai-gateway",
      "name": "Vercel AI Gateway",
      "company": "Vercel Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "Depends on the routed model; rate limits aren't published, the free tier has lower per-model limits, and a 402 quota_for_entity_exceeded stops a spent budget.",
      "source": "https://vercel.com/docs/ai-gateway/rate-limits.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/vercel-ai-gateway"
    },
    {
      "slug": "vertex-ai",
      "name": "Vertex AI (Gemini Enterprise Agent Platform)",
      "company": "Google LLC",
      "grade": "AA",
      "fit": "caveat",
      "why": "Gemini ships a 1M-token window and thinking tokens share it under a soft budget; Standard PayGo has no fixed quota but shared-capacity 429s that acc-26 could not clear with 30 minutes of backoff; Claude has per-model QPM buckets.",
      "source": "https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/long-context",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/vertex-ai"
    },
    {
      "slug": "xai-api",
      "name": "xAI API (Grok)",
      "company": "SpaceXSI (xAI)",
      "grade": "BB",
      "fit": "caveat",
      "why": "Grok 4.7 prices double above 200K prompt tokens for the whole request, and TPM caps by spend tier (Tier 0 at $0) can refuse a 300K request; context sizes are on the models page.",
      "source": "https://docs.x.ai/developers/models.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/xai-api"
    },
    {
      "slug": "antseed",
      "name": "Antseed",
      "company": "Antseed Foundation",
      "grade": "C",
      "fit": "unchecked",
      "why": "Context depends on the seller's model; the buyer waits 5 minutes for a non-streaming response by default and nothing states accepted input sizes.",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/antseed"
    },
    {
      "slug": "cloudflare-workers-ai",
      "name": "Cloudflare Workers AI",
      "company": "Cloudflare, Inc.",
      "grade": "BB",
      "fit": "unchecked",
      "why": "Context lengths live on each model page, none of which were read; the limits page lists request rates only.",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cloudflare-workers-ai"
    },
    {
      "slug": "novita-ai",
      "name": "Novita AI",
      "company": "Novita AI",
      "grade": "C",
      "fit": "unchecked",
      "why": "Context lengths live on model pages drawn by script; the chat reference gives no context figure and per-model rate limits weren't readable.",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/novita-ai"
    },
    {
      "slug": "simplismart",
      "name": "Simplismart",
      "company": "Verute Technologies Private Limited",
      "grade": "F",
      "fit": "unchecked",
      "why": "Context lengths aren't documented for shared endpoints.",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/simplismart"
    },
    {
      "slug": "stepfun-api",
      "name": "StepFun API",
      "company": "STEPFUN TECHNOLOGY (HK) LIMITED",
      "grade": "E",
      "fit": "unchecked",
      "why": "The host platform.stepfun.ai closed the connection on every attempt (curl exit 56, no HTTP status) from this session, so no page was read.",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/stepfun-api"
    },
    {
      "slug": "tencent-tokenhub",
      "name": "Tencent Cloud TokenHub",
      "company": "Tencent Cloud Computing (Beijing) Co., Ltd. (腾讯云计算（北京）有限责任公司)",
      "grade": "C",
      "fit": "unchecked",
      "why": "The host cloud.tencent.com closed the connection on every attempt (curl exit 56, no HTTP status) from this session, so no page was read.",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/tencent-tokenhub"
    },
    {
      "slug": "cerebras-inference",
      "name": "Cerebras Inference",
      "company": "Cerebras Systems Inc.",
      "grade": "B",
      "fit": "no",
      "why": "Shared inference models stop at 65,536 (Qwen 3.8 27B) or 131,072 (GPT OSS 120B) tokens on the Hugging Face router's listing, so a 300K request is not accepted.",
      "source": "https://router.huggingface.co/v1/models",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/cerebras-inference"
    },
    {
      "slug": "cohere-chat",
      "name": "Cohere Chat API (Command models)",
      "company": "Cohere Inc.",
      "grade": "B",
      "fit": "no",
      "why": "Command A's context is 256K and the trial key allows 100,000 tokens a minute, so a 300K-token request is not accepted.",
      "source": "https://docs.cohere.com/docs/models",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cohere-chat"
    },
    {
      "slug": "elorian-glimpse",
      "name": "Glimpse",
      "company": "Elorian AI, Inc.",
      "grade": "F",
      "fit": "no",
      "why": "Context length is 262,144 tokens, below a 300K request.",
      "source": "https://openrouter.ai/api/v1/models/elorian/glimpse/endpoints",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/elorian-glimpse"
    },
    {
      "slug": "groq",
      "name": "GroqCloud",
      "company": "Groq LLC",
      "grade": "BB",
      "fit": "no",
      "why": "Every self-serve model is served at 131,072 tokens of context (Hugging Face router listing for gpt-oss), so a 300K request is not accepted.",
      "source": "https://router.huggingface.co/v1/models",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/groq"
    },
    {
      "slug": "mistral-api",
      "name": "Mistral AI API",
      "company": "Mistral AI (RCS Paris 952 418 325)",
      "grade": "B",
      "fit": "no",
      "why": "The largest context listed is 256K tokens and requests over the window return 400, so a 300K-token request can't be accepted.",
      "source": "https://docs.mistral.ai/resources/known-limitations.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/mistral-api"
    },
    {
      "slug": "swe-2-flash",
      "name": "SWE-2 Flash",
      "company": "Cognition AI, Inc.",
      "grade": "F",
      "fit": "no",
      "why": "No context length published and no API.",
      "source": "https://docs.devin.ai/desktop/models.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/swe-2-flash"
    }
  ],
  "marks": "yes: the docs show it does the job as the success test says. caveat: it does the job with a limit an agent should know. no: it doesn't do this job. unchecked: we couldn't read it.",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "useCase": {
    "accounts": 3,
    "after": [
      "inference.structured-extraction"
    ],
    "area": "models",
    "asks": [
      "load the entire repo and answer questions about it",
      "read these 300K tokens of history and summarise",
      "analyse this legacy monolith before we split it"
    ],
    "before": [
      "web-scrapers.crawl-site",
      "document-extraction.pdf-to-text"
    ],
    "blockers": [
      "free or low tier tokens-per-minute cap below the context size, so a single large request is a 429",
      "the 1M window is shared with reasoning and output, so a long input leaves little room to answer",
      "context preparation takes longer than the call"
    ],
    "capabilities": [
      "inference.llm"
    ],
    "category": "inference",
    "categoryName": "Model APIs \u0026 inference",
    "checked": "2026-11-12",
    "companies": {
      "fit": 4,
      "fitWithLimit": 23,
      "tagged": 42
    },
    "competition": "crowded",
    "fits": [
      "anthropic-api",
      "deepinfra",
      "deepseek-api",
      "meta-model-api"
    ],
    "fitsWithLimit": [
      "amazon-bedrock",
      "azure-openai",
      "blockrun-ai",
      "cloudflare-ai-gateway",
      "eden-ai",
      "fireworks-inference",
      "gemini-api",
      "hugging-face-inference-providers",
      "jais-3",
      "litellm",
      "minimax-api",
      "muse-spark-open",
      "nemotron-3-5-super-vl",
      "openai-api",
      "openrouter",
      "prism-inference",
      "ramp-router",
      "reflection-beam",
      "routerplus",
      "sambanova",
      "siliconflow",
      "together-inference",
      "vercel-ai-gateway",
      "vertex-ai",
      "xai-api"
    ],
    "id": "inference.long-context-read",
    "listings": {
      "caveat": 25,
      "no": 6,
      "notRead": 6,
      "tagged": 51,
      "unchecked": 6,
      "yes": 4
    },
    "name": "Read a whole codebase or corpus in one call",
    "status": "draft",
    "success": "A 300K-token request is accepted on the tier the caller is on. Measured: it answers in under two minutes, and the response cites content from the end of the prompt as well as the start.",
    "url": "https://www.anchorterminal.com/api/v1/use-cases/inference.long-context-read.json",
    "weight": 8,
    "weightBasis": "3 of 28 accounts; one team measured that 88 per cent of sessions used under 200K tokens, so the job is real but occasional"
  }
}
