AI Model ID Lookup
Find the exact API model ID for GPT, Claude, Gemini, Grok, DeepSeek, and Mistral models
| Model | Provider | Model ID | Alias | Reasoning | Input / 1M | Output / 1M |
|---|---|---|---|---|---|---|
| GPT-5.6 Sol | OpenAI | - | none / low / medium / high / xhigh / max | $4 | $20 | |
| GPT-5.6 Luna | OpenAI | - | none / low / medium / high / xhigh / max | $0.2 | $1.2 | |
| GPT-5.6 Terra | OpenAI | - | none / low / medium / high / xhigh / max | $2 | $12 | |
| GPT-5.6 | OpenAI | - | none / low / medium / high / xhigh / max | $4 | $20 | |
| GPT-5.5 | OpenAI | - | none / low / medium / high / xhigh | $5 | $30 | |
| GPT-5.4 | OpenAI | - | none / low / medium / high / xhigh | $2.5 | $15 | |
| GPT-5.4 mini | OpenAI | - | none / low / medium / high / xhigh | $0.75 | $4.5 | |
| GPT-5.4 nano | OpenAI | - | none / low / medium / high / xhigh | $0.2 | $1.25 | |
| GPT-5.2 | OpenAI | - | none / low / medium / high / xhigh | $1.75 | $14 | |
| GPT-5.1 | OpenAI | - | none / low / medium / high | $1.25 | $10 | |
| GPT-5 | OpenAI | - | minimal / low / medium / high | $1.25 | $10 | |
| GPT-5 Mini | OpenAI | - | minimal / low / medium / high | $0.25 | $2 | |
| GPT-5 Nano | OpenAI | - | minimal / low / medium / high | $0.05 | $0.4 | |
| o3 | OpenAI | - | low / medium / high | $2 | $8 | |
| GPT-4o | OpenAI | - | - | $2.5 | $10 | |
| GPT-4o mini | OpenAI | - | - | $0.15 | $0.6 | |
| GPT-4.1 | OpenAI | - | - | $2 | $8 | |
| GPT-4.1 mini | OpenAI | - | - | $0.4 | $1.6 | |
| Claude Opus 5 | Anthropic | - | low / medium / high / xhigh / max | $5 | $25 | |
| Claude Sonnet 5 | Anthropic | - | - | $2 | $10 | |
| Claude Fable 5 | Anthropic | - | low / medium / high / xhigh / max | $10 | $50 | |
| Claude Opus 4.8 | Anthropic | - | low / medium / high / xhigh / max | $5 | $25 | |
| Claude Opus 4.7 | Anthropic | - | low / medium / high / xhigh / max | $5 | $25 | |
| Claude Opus 4.6 | Anthropic | - | low / medium / high / max | $5 | $25 | |
| Claude Opus 4.5 | Anthropic | low / medium / high | $5 | $25 | ||
| Claude Sonnet 4.6 | Anthropic | - | low / medium / high / max | $3 | $15 | |
| Claude Sonnet 4.5 | Anthropic | - | $3 | $15 | ||
| Claude Haiku 4.5 | Anthropic | - | $1 | $5 | ||
| Gemini 3.7 Flash | - | low / medium / high | $0.75 | $3.75 | ||
| Gemini 3.6 Flash | - | minimal / low / medium / high | $0.75 | $3.75 | ||
| Gemini 3.5 Flash | - | minimal / low / medium / high | $1.5 | $9 | ||
| Gemini 3.5 Flash Lite | - | minimal / low / medium / high | $0.3 | $2.5 | ||
| Gemini 3.1 Pro Preview | - | low / medium / high | $2 | $12 | ||
| Gemini 3.1 Flash Lite | - | minimal / low / medium / high | $0.25 | $1.5 | ||
| Gemini 3 Flash Preview | - | minimal / low / medium / high | $0.5 | $3 | ||
| Gemini 2.5 Pro | - | - | $1.25 | $10 | ||
| Gemini 2.5 Flash | - | - | $0.3 | $2.5 | ||
| Gemini 2.5 Flash-Lite | - | - | $0.1 | $0.4 | ||
| Grok 4.6 | xAI | - | low / medium / high / xhigh | $2 | $6 | |
| Grok 4.5 | xAI | - | low / medium / high | $2 | $6 | |
| Grok 4.3 | xAI | - | none / low / medium / high | $1.25 | $2.5 | |
| Grok 4.20 Multi-Agent | xAI | - | low / medium / high / xhigh | $1.25 | $2.5 | |
| Grok 4.20 | xAI | - | - | $1.25 | $2.5 | |
| DeepSeek V4 Pro | DeepSeek | - | - | $0.435 | $0.87 | |
| DeepSeek V4 Flash | DeepSeek | - | - | $0.14 | $0.28 | |
| Mistral Large 3 | Mistral | - | $0.5 | $1.5 | ||
| Mistral Medium 3.5 | Mistral | none / high | $1.5 | $7.5 | ||
| Mistral Small 4 | Mistral | none / high | $0.15 | $0.6 | ||
| Magistral Medium (latest) | Mistral | - | - | $2 | $5 | |
| Codestral (latest) | Mistral | - | - | $0.3 | $0.9 | |
| Devstral Medium | Mistral | - | $0.4 | $2 |
About AI Model IDs
A model ID is the literal string a provider's API expects in the model field of a request - not the display name used in announcements or pricing pages. Passing the wrong one is one of the most common causes of a failed API call.
Note: providers rename, deprecate, and re-price models often - confirm against the provider's own docs before relying on this for production code.
Every provider names its models differently in marketing copy than in the string you actually pass to the API. This page is a searchable reference for the exact model ID, any rolling alias, reasoning-effort support, and per-1M-token pricing for the current GPT, Claude, Gemini, Grok, DeepSeek, and Mistral lineups - all bundled into the page, so searching and copying never leaves your browser.
How to use it
- Type a model name, ID, or provider into the search box, or click a provider filter.
- Find the row for the model you want to call from your code.
- Click the copy icon next to the Model ID (or Alias) to copy it to your clipboard.
- Check the Reasoning column to see whether that model accepts an effort parameter, and the price columns for current per-1M-token cost.
Frequently asked questions
What's the difference between a model name and a model ID?
The name (like "Claude Opus 5") is what a provider calls a model in its marketing and docs. The ID (like claude-opus-5) is the exact string you pass to that provider's API - get it wrong and the request fails with a 404 or "model not found" error.
What does the Alias column mean?
Some providers publish a rolling alias alongside a dated, pinned model ID - for example claude-haiku-4-5 always points at the latest snapshot, while claude-haiku-4-5-20251001 pins to that exact version forever. Use the alias for convenience, the dated ID when you need reproducible behavior.
What are the reasoning effort levels?
Newer reasoning-capable models let you trade latency and cost for answer quality via an effort parameter (e.g. low/medium/high, sometimes up to xhigh or max). A model with no reasoning column entry doesn't expose this control - it reasons the same way on every call.
Where does this data come from, and does it send anything anywhere?
The list is a hand-curated snapshot bundled into this page - lookups and filtering all run in your browser against that fixed dataset, with no API calls or tracking. Provider pricing and lineups change often, so double-check the provider's own docs before shipping to production.
Why isn't model X on the list?
This covers the current flagship and commonly-used models from OpenAI, Anthropic, Google, xAI, DeepSeek, and Mistral. Older deprecated snapshots and narrow regional variants are left out to keep the list scannable.