latestmodel.id
Proprietary resolver service · no signup, no key

Stop hardcoding model ids.

One API. Reference a tag like smartest, cheapest, or fastest — get back the live model id at the provider you choose. We track deprecations so your code doesn't.

Explore modelsAPI docs
  • 223
    Models
  • 7
    Providers
  • 5 min
    Edge cache
  • 99.9%
    Uptime SLO
Live registry·tag → model id
  • smartest
    claude-opus-5-5
    highest intelligence
  • cheapest
    gemma-4-26b-a4b-it
    lowest input price
  • fastest
    openai/gpt-oss-20b
    highest tok/s
  • longest_context
    gpt-6-astra
    largest window
GEThttps://latestmodel.id/v1/resolve?tag=smartest&format=string
One resolver — 7 frontier providers · 223 models · 4 gateways
Anthropic21 models
DeepSeek6 models
Google56 models
Groq20 models
Mistral AI23 models
OpenAI70 models
xAI27 models
AWS Bedrock17 models
Azure OpenAI2 models
GCP Vertex24 models
OpenRouter69 models
FireworksComing soon
TogetherComing soon
Try it live

Resolve a tag

Pick a tag and a provider, see the resolved model id update in real time. Copy the URL to use it from your code, or share the demo state with a teammate.

smartest — Highest intelligence_index among GA chat/reasoning models.

GEThttps://latestmodel.id/api/v1/resolve?tag=smartest&format=string
→…
Live

Real-time analytics

0 origin requests · top picks across the resolver. CDN cache hits aren't counted — this is what hit our origin.

Today
0
7-day
0
30-day
0
Lifetime
0
Waiting for the first request

Once traffic arrives, this panel shows top tags, providers, models, and types in real time. Counters refresh every 30 seconds.

Explorer

Filter, sort, and compare every model in the registry. Your filter state is encoded into the URL — copy it once, share it anywhere.

65 of 223 models
Compare
Pick any two metrics. One dot per model, color by provider.
65 models
claude-fable-5-1
Claude Fable 5.1
anthropicchat1.0M$10.00$50.00——
claude-mythos-5-1preview
Claude Mythos 5.1
anthropicchat1.0M$10.00$50.00——
claude-opus-5-5
Claude Opus 5.5
anthropicchat1.0M$4.00$20.00——
claude-opus-5
Claude Opus 5
anthropicchat1.0M$5.00$25.00——
claude-sonnet-5
Claude Sonnet 5
anthropicchat1.0M$2.00$10.00——
deepseek-flash
DeepSeek V4.1 Flash
deepseekchat1.0M$0.150$0.600——
gemini-3.6-flash
Gemini 3.6 Flash
googlechat1.0M$0.750$3.75——
gemini-3.7-flash
Gemini 3.7 Flash
googlechat1.0M$0.750$3.75——
gemini-3.8-flash
Gemini 3.8 Flash
googlechat1.0M$0.750$3.75——
gemini-3.5-flash-lite
Gemini 3.5 Flash-Lite
googlechat1.0M$0.300$2.50350—
gemini-2.5-computer-use-preview-10-2025preview
Gemini 2.5 Computer Use Preview
googlechat1.0M$1.25$10.00——
gemma-4-26b-a4b-it
Gemma 4 26B A4B (Instruction-Tuned)
googlechat262Kfreefree——
gemma-4-31b-it
Gemma 4 31B (Instruction-Tuned)
googlechat262Kfreefree——
gemini-robotics-er-2-previewpreview
Gemini Robotics-ER 2 (Preview)
googlechat131K$1.00$5.00——
deep-research-pro-preview-12-2025preview
Deep Research Pro (Preview)
googlechat1.0M————
deep-research-preview-04-2026preview
Deep Research (Preview)
googlechat1.0M————
deep-research-max-preview-04-2026preview
Deep Research Max (Preview)
googlechat1.0M————
antigravity-preview-09-2026preview
Antigravity Agent (Preview, September 2026)
googlechat1.0M————
openai/gpt-oss-safeguard-20bpreview
GPT-OSS Safeguard 20B (Groq)
groqchat131K$0.075$0.3001000—
minimaxai/minimax-m2.7preview
MiniMax M2.7 (Groq)
groqchat197K——260—
qwen/qwen3.8-27bpreview
Qwen 3.8 27B (Groq)
groqchat131K$0.800$4.00450—
zai-glm-5-2preview
Z.ai GLM 5.2
mistralchat1.0M$1.40$4.40——
zai-glm-5-3preview
Z.ai GLM 5.3
mistralchat1.0M$1.40$4.40——
gpt-6-astra
GPT-6 Astra
openaichat1.1M$10.00$50.00——
gpt-6-sol
GPT-6 Sol
openaichat1.1M$2.00$10.00——
gpt-6-luna
GPT-6 Luna
openaichat1.1M$0.100$0.500——
gpt-5.6-sol
GPT-5.6 Sol
openaichat1.1M$4.00$20.00——
gpt-5.6-terra
GPT-5.6 Terra
openaichat1.1M$2.00$12.00——
gpt-5.6-luna
GPT-5.6 Luna
openaichat1.1M$0.200$1.20——
gpt-5.1
GPT-5.1
openaichat400K$1.25$10.00——
gpt-5.2
GPT-5.2
openaichat400K$1.75$14.00——
gpt-4.1-mini
GPT-4.1 Mini
openaichat1.0M$0.400$1.60——
gpt-4o
GPT-4o
openaichat128K$2.50$10.00——
gpt-4o-mini
GPT-4o Mini
openaichat128K$0.150$0.600——
grok-build-0.1
Grok Build 0.1
xaichat256K$1.00$2.00——
claude-mythos-5preview
Claude Mythos 5
anthropicchat1.0M$10.00$50.00—92
claude-fable-5
Claude Fable 5
anthropicchat1.0M$10.00$50.00—90
gpt-5.5
GPT-5.5
openaichat1.1M$5.00$30.00—88
gemini-3.5-flash
Gemini 3.5 Flash
googlechat1.0M$1.50$9.00—86
claude-opus-4-8
Claude Opus 4.8
anthropicchat1.0M$5.00$25.00—85
gemini-3.1-pro-previewpreview
Gemini 3.1 Pro
googlechat1.0M$2.00$12.00—84
claude-opus-4-7
Claude Opus 4.7
anthropicchat1.0M$5.00$25.00—82
gpt-5.4
GPT-5.4
openaichat1.0M$2.50$15.00—80
claude-opus-4-6
Claude Opus 4.6
anthropicchat1.0M$5.00$25.00—78
claude-opus-4-5-20251101
Claude Opus 4.5
anthropicchat200K$5.00$25.00—75
gemini-2.5-pro
Gemini 2.5 Pro
googlechat1.0M$1.25$10.00—75
claude-sonnet-4-6
Claude Sonnet 4.6
anthropicchat1.0M$3.00$15.00—73
claude-sonnet-4-5-20250929
Claude Sonnet 4.5
anthropicchat200K$3.00$15.00—70
gemini-3-flash-previewpreview
Gemini 3 Flash
googlechat1.0M$0.500$3.0025070
mistral-medium-3-5
Mistral Medium 3.5
mistralchat262K$1.50$7.50—70
gpt-5.4-mini
GPT-5.4 Mini
openaichat400K$0.750$4.5013070
grok-4.20-0309-non-reasoning
Grok 4.20 (Non-Reasoning)
xaichat1.0M$1.25$2.5010070
mistral-large-2512
Mistral Large 3
mistralchat262K$0.500$1.50—68
gemini-2.5-flash
Gemini 2.5 Flash
googlechat1.0M$0.300$2.5025065
openai/gpt-oss-120b
GPT-OSS 120B (Groq)
groqchat131K$0.150$0.60050065
claude-haiku-4-5-20251001
Claude Haiku 4.5
anthropicchat200K$1.00$5.0013060
gemini-3.1-flash-lite
Gemini 3.1 Flash-Lite
googlechat1.0M$0.250$1.5030060
mistral-small-2603
Mistral Small 4
mistralchat262K$0.150$0.600—60
gpt-4.1
GPT-4.1
openaichat1.0M$2.00$8.00—60
gemini-2.5-flash-lite
Gemini 2.5 Flash-Lite
googlechat1.0M$0.100$0.40035055
gpt-5.4-nano
GPT-5.4 Nano
openaichat400K$0.200$1.2520055
openai/gpt-oss-20b
GPT-OSS 20B (Groq)
groqchat131K$0.075$0.300100050
ministral-14b-2512
Ministral 3 14B
mistralchat262K$0.200$0.200—50
ministral-8b-2512
Ministral 3 8B
mistralchat262K$0.150$0.150—45
ministral-3b-2512
Ministral 3 3B
mistralchat262K$0.100$0.100—38
GET/v1/resolveResolve a tag to a model id. The killer endpoint.
  • https://latestmodel.id/v1/resolve?tag=smartest&provider=anthropic&format=string
  • https://latestmodel.id/v1/resolve?tag=cheapest&type=chat
  • https://latestmodel.id/v1/resolve?tag=fastest&min_context=100000&supports=tools,vision
  • https://latestmodel.id/v1/resolve?tag=smartest&aggregator=openrouter
  • https://latestmodel.id/v1/resolve?tag=smartest_embedding&type=embedding
GET/v1/modelsList all models with filters and sorting.
  • https://latestmodel.id/v1/models?provider=anthropic
  • https://latestmodel.id/v1/models?type=image_generation
  • https://latestmodel.id/v1/models?supports=tools&min_context=100000
  • https://latestmodel.id/v1/models?sort=performance.intelligence_index&order=desc&limit=10
  • https://latestmodel.id/v1/models?where=status%20%3D%3D%20%27ga%27%20%26%26%20capabilities.vision%20%3D%3D%20true
GET/v1/models/{id}Single model detail.
GET/v1/providersList of providers and counts.
GET/v1/tagsList of tag definitions.
GET/v1/healthService status + provider health.

Response shape

{
  "data": <payload>,
  "meta": {
    "version": "...",
    "as_of": "2026-05-08T12:00:00Z",
    "request_id": "..."
  },
  "errors": []
}

Append &format=string to /v1/resolve to get a plain-text body with just the model id.

Intelligence index

A 0–100 composite from public benchmarks: MMLU-Pro, GPQA Diamond, HumanEval, ARC-AGI v2, and Arena Elo (LMArena), normalized and weighted. We do not run our own evaluations — every score is cited back to its source so you can audit it via the model drawer.

Pricing

Taken from each provider's official pricing page on the date in last_verified. Multi-modal pricing is split into typed fields (input, output, cached, batch, per image, per minute, per second).

Tags

Tags are computed at query time from the catalog using a small expression DSL: paths, comparisons, and boolean operators. The full tag definitions are exposed at /v1/tags.

Updates

A daily automated routine pings each provider's /models endpoint, diffs against the registry, and queues changes for review. A human approves every update before it ships — no silent edits.

Caching

Edge cache: s-maxage=300, stale-while-revalidate=86400. Even if our origin is down for 24h, Cloudflare keeps serving stale responses, so your app never sees a hard failure.