LLM Model ID Reference
The exact model ID string for every major LLM API — copy-paste ready, with context windows and prices, verified against official docs.
Model IDs verified 2026-07-24 against official provider docs. Click an ID to copy it.
OpenAI (5)Anthropic Claude (5)Google Gemini (7)xAI Grok (2)DeepSeek (2)Mistral AI (4)Groq (3)
OpenAI model IDs
| Model | API model ID | Context | $/1M in · out |
|---|---|---|---|
| GPT-5.6 Sol | — | $5 · $30 | |
| GPT-5.6 Terra | — | $2.5 · $15 | |
| GPT-5.6 Luna | — | $1 · $6 | |
| GPT-5.4 mini | — | $0.75 · $4.5 | |
| GPT-5.4 nano | — | $0.2 · $1.25 |
Source: official OpenAI docs → · OpenAI cost calculator · OpenAI model IDs page
Anthropic Claude model IDs
| Model | API model ID | Context | $/1M in · out |
|---|---|---|---|
| Claude Opus 5 | 1,000,000 | $5 · $25 | |
| Claude Opus 4.8 | 1,000,000 | $5 · $25 | |
| Claude Sonnet 5 | 1,000,000 | $2 · $10 | |
| Claude Haiku 4.5 | — | $1 · $5 | |
| Claude Fable 5 | — | $10 · $50 |
Source: official Anthropic Claude docs → · Anthropic Claude cost calculator · Anthropic Claude model IDs page
Google Gemini model IDs
| Model | API model ID | Context | $/1M in · out |
|---|---|---|---|
| Gemini 3.6 Flash | — | $1.5 · $7.5 | |
| Gemini 3.5 Flash | — | $1.5 · $9 | |
| Gemini 3.5 Flash-Lite | — | $0.3 · $2.5 | |
| Gemini 3.1 Pro | — | $2 · $12 | |
| Gemini 2.5 Pro | 1,000,000 | $1.25 · $10 | |
| Gemini 2.5 Flash | 1,000,000 | $0.3 · $2.5 | |
| Gemini 2.5 Flash-Lite | — | $0.1 · $0.4 |
Source: official Google Gemini docs → · Google Gemini cost calculator · Google Gemini model IDs page
xAI Grok model IDs
Source: official xAI Grok docs → · xAI Grok cost calculator · xAI Grok model IDs page
DeepSeek model IDs
| Model | API model ID | Context | $/1M in · out |
|---|---|---|---|
| DeepSeek V4 Flash | 1,000,000 | $0.14 · $0.28 | |
| DeepSeek V4 Pro | 1,000,000 | $0.435 · $0.87 |
Source: official DeepSeek docs → · DeepSeek cost calculator · DeepSeek model IDs page
Mistral AI model IDs
| Model | API model ID | Context | $/1M in · out |
|---|---|---|---|
| Mistral Large 3 | — | $0.5 · $1.5 | |
| Mistral Medium 3.5 | — | $1.5 · $7.5 | |
| Mistral Small 4 | — | $0.15 · $0.6 | |
| Ministral 3 3B | — | $0.1 · $0.1 |
Source: official Mistral AI docs → · Mistral AI cost calculator · Mistral AI model IDs page
Groq model IDs
| Model | API model ID | Context | $/1M in · out |
|---|---|---|---|
| Llama 3.3 70B Versatile | 131,072 | $0.59 · $0.79 | |
| Llama 3.1 8B Instant | 131,072 | $0.05 · $0.08 | |
| GPT-OSS 120B | 131,072 | $0.15 · $0.6 |
Source: official Groq docs → · Groq cost calculator · Groq model IDs page
Why a model ID reference
Every LLM API call names its model with an exact string — the model field — and getting it slightly wrong returns a 404 or an "unknown model" error, not a helpful correction. The strings are easy to get wrong because marketing names and API IDs differ in ways that don't follow one rule: Claude Opus 4.8 is claude-opus-4-8 (dashes, not dots), Gemini 3.1 Pro is gemini-3.1-pro-preview (a suffix the product name doesn't carry), GPT-OSS 120B on Groq is openai/gpt-oss-120b (a vendor prefix), and Mistral's current models are called via -latest aliases like mistral-large-latest.
This page lists the verified, current ID for every model in one place. Click any ID to copy it.
How to use it
- Find your provider and model in the tables above.
- Click the model ID to copy the exact string.
- Paste it into the
modelfield of your API request or SDK call.
Each entry also shows the context window and the input/output price per 1M tokens, so you can sanity-check you're calling the model you think you're calling. To estimate what a workload costs on any of these models, use the LLM cost calculator; to check that your key works, use the API key tester.
How current is this?
The data behind this page is verified against each provider's official pricing and model-catalog pages, and the verification date is shown above. Providers rename, deprecate, and retire models frequently — when an ID stops appearing here, it usually means the provider retired it, and calls using it will start failing.
Related tools
Building with AI APIs? Elegant Atomics turns AI capability into B2B SaaS growth. Work with us →