modelparams.dev
API GitHub

Providers

Alibaba Anthropic Cohere DeepSeek Google Meta MiniMax Mistral Moonshot AI Nvidia OpenAI Perplexity xAI Xiaomi Z.ai

Providers

Alibaba Anthropic Cohere DeepSeek Google Meta MiniMax Mistral Moonshot AI Nvidia OpenAI Perplexity xAI Xiaomi Z.ai
Length integer 127 models

Max tokens max_tokens

Maximum number of output tokens the model may generate.

Type
integer
Default
varies by model
Range
1 – 128000

max_tokens by model

The exact type, default, range, and conditions for max_tokens on each model that documents it.

Model Provider Type Default Range / values Condition
Qwen Flash Alibaba integer — 1…+∞ —
Qwen Plus Alibaba integer — 1…+∞ —
Qwen3 Coder Flash Alibaba integer — 1…+∞ —
Qwen3 Coder Plus Alibaba integer — 1…+∞ —
Qwen3 Max Alibaba integer — 1…+∞ —
Qwen3.5 Alibaba integer — 1…+∞ —
Qwen3.5 Flash Alibaba integer — 1…+∞ —
Qwq Plus Alibaba integer — 1…+∞ —
Claude 3 Opus 20240229 Anthropic integer 4096 1…+∞ —
Claude 3 Opus Latest Anthropic integer 4096 1…+∞ —
Claude 3.5 Haiku 20241022 Anthropic integer 4096 1…+∞ —
Claude 3.5 Haiku Latest Anthropic integer 4096 1…+∞ —
Claude 3.5 Sonnet 20241022 Anthropic integer 4096 1…+∞ —
Claude 3.5 Sonnet Latest Anthropic integer 4096 1…+∞ —
Claude 3.7 Sonnet 20250219 Anthropic integer 4096 1…+∞ —
Claude 3.7 Sonnet Latest Anthropic integer 4096 1…+∞ —
Claude Fable 5 Anthropic integer 4096 1…+∞ —
Claude Fable 5 Subscription Anthropic integer 4096 1…+∞ —
Claude Haiku 4 Anthropic integer 4096 1…+∞ —
Claude Haiku 4 Subscription Anthropic integer 4096 1…+∞ —
Claude Haiku 4.5 Anthropic integer 4096 1…+∞ —
Claude Haiku 4.5 Subscription Anthropic integer 4096 1…+∞ —
Claude Haiku 4.5 20251001 Anthropic integer 4096 1…+∞ —
Claude Haiku 4.5 20251001 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4 20250514 Anthropic integer 4096 1…+∞ —
Claude Opus 4 20250514 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4.1 20250805 Anthropic integer 4096 1…+∞ —
Claude Opus 4.1 20250805 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4.5 20251101 Anthropic integer 4096 1…+∞ —
Claude Opus 4.5 20251101 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4.6 Anthropic integer 4096 1…+∞ —
Claude Opus 4.6 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4.7 Anthropic integer 4096 1…+∞ —
Claude Opus 4.7 Subscription Anthropic integer 4096 1…+∞ —
Claude Opus 4.8 Anthropic integer 4096 1…+∞ —
Claude Opus 4.8 Subscription Anthropic integer 4096 1…+∞ —
Claude Sonnet 4 Subscription Anthropic integer 4096 1…+∞ —
Claude Sonnet 4 20250514 Anthropic integer 4096 1…+∞ —
Claude Sonnet 4 20250514 Subscription Anthropic integer 4096 1…+∞ —
Claude Sonnet 4.5 Anthropic integer 4096 1…+∞ —
Claude Sonnet 4.5 Subscription Anthropic integer 4096 1…+∞ —
Claude Sonnet 4.5 20250929 Anthropic integer 4096 1…+∞ —
Claude Sonnet 4.5 20250929 Subscription Anthropic integer 4096 1…+∞ —
Claude Sonnet 4.6 Anthropic integer 4096 1…+∞ —
Claude Sonnet 4.6 Subscription Anthropic integer 4096 1…+∞ —
Command A 03 2025 Cohere integer — 1…+∞ —
Command A Plus 05 2026 Cohere integer — 1…+∞ —
Command A Reasoning 08 2025 Cohere integer — 1…+∞ —
Command A Translate 08 2025 Cohere integer — 1…+∞ —
Command A Vision 07 2025 Cohere integer — 1…+∞ —
Command R 08 2024 Cohere integer — 1…+∞ —
Command R Plus 08 2024 Cohere integer — 1…+∞ —
Command R7b 12 2024 Cohere integer — 1…+∞ —
Deepseek Chat DeepSeek integer 4096 1…+∞ —
Deepseek Reasoner DeepSeek integer 4096 1…+∞ —
Deepseek V4 Flash DeepSeek integer 4096 1…+∞ —
Deepseek V4 Pro DeepSeek integer 4096 1…+∞ —
MiniMax M2 Subscription MiniMax integer — 1…+∞ —
MiniMax M2.1 Subscription MiniMax integer — 1…+∞ —
MiniMax M2.1 Highspeed Subscription MiniMax integer — 1…+∞ —
MiniMax M2.5 Subscription MiniMax integer — 1…+∞ —
MiniMax M2.5 Highspeed Subscription MiniMax integer — 1…+∞ —
MiniMax M2.7 Subscription MiniMax integer — 1…+∞ —
MiniMax M2.7 Highspeed Subscription MiniMax integer — 1…+∞ —
MiniMax M3 Subscription MiniMax integer — 1…+∞ —
Codestral Latest Mistral integer — 1…+∞ —
Devstral 2512 Mistral integer — 1…+∞ —
Devstral Latest Mistral integer — 1…+∞ —
Magistral Medium Latest Mistral integer — 1…+∞ —
Magistral Small Latest Mistral integer — 1…+∞ —
Ministral 14b Latest Mistral integer — 1…+∞ —
Ministral 3b Latest Mistral integer — 1…+∞ —
Ministral 8b Latest Mistral integer — 1…+∞ —
Mistral Large Latest Mistral integer — 1…+∞ —
Mistral Medium 3.5 Mistral integer — 1…+∞ —
Mistral Medium Latest Mistral integer — 1…+∞ —
Mistral Small Latest Mistral integer — 1…+∞ —
Open Mistral Nemo Mistral integer — 1…+∞ —
Llama 3.1 Nemoguard 8b Topic Control Nvidia integer 1024 1…+∞ —
Llama 3.1 Nemotron Nano 8b V1 Nvidia integer 4096 1…16384 —
Llama 3.1 Nemotron Ultra 253b V1 Nvidia integer 4096 1…16384 —
Llama 3.3 Nemotron Super 49b V1 Nvidia integer 4096 1…16384 —
Llama 3.3 Nemotron Super 49b V1.5 Nvidia integer 65536 1…65536 —
Nemotron 3 Nano 30b A3b Nvidia integer 16384 1…32768 —
Nemotron 3 Super 120b A12b Nvidia integer 16384 1…32768 —
Nemotron 3 Ultra Subscription Nvidia integer 16384 1…32768 —
Nemotron 3 Ultra 550b A55b Nvidia integer 16384 1…32768 —
Nemotron Content Safety Reasoning 4b Nvidia integer 16384 1…32768 —
Nemotron Mini 4b Instruct Nvidia integer 1024 1…4096 —
Riva Translate 4b Instruct V1.1 Nvidia integer 512 1…4096 —
Usdcode Llama 3.1 70b Instruct Nvidia integer 1024 1…2048 —
Chatgpt 4o Latest OpenAI integer 4096 1…+∞ —
Gpt 3.5 Turbo OpenAI integer 4096 1…+∞ —
Gpt 4 Turbo OpenAI integer 4096 1…+∞ —
Gpt 4 Turbo 2024-04-09 OpenAI integer 4096 1…+∞ —
Gpt 4.1 OpenAI integer 4096 1…+∞ —
Gpt 4.1 Mini OpenAI integer 4096 1…+∞ —
Gpt 4.1 Nano OpenAI integer 4096 1…+∞ —
Gpt 4o 2024-11-20 OpenAI integer 4096 1…+∞ —
GPT-4o OpenAI integer 4096 1…+∞ —
GPT-4o mini OpenAI integer 4096 1…+∞ —
O1 Preview OpenAI integer 4096 1…+∞ —
o1-mini OpenAI integer 4096 1…+∞ —
Sonar Perplexity integer — 1…128000 —
Sonar Deep Research Perplexity integer — 1…128000 —
Sonar Pro Perplexity integer — 1…128000 —
Sonar Reasoning Pro Perplexity integer — 1…128000 —
GLM-4.5 Z.ai integer — 1…+∞ —
GLM-4.5 Subscription Z.ai integer — 1…+∞ —
GLM-4.5-Air Z.ai integer — 1…+∞ —
GLM-4.5-Air Subscription Z.ai integer — 1…+∞ —
GLM-4.5-AirX Z.ai integer — 1…+∞ —
GLM-4.5-Flash Z.ai integer — 1…+∞ —
GLM-4.5-X Z.ai integer — 1…+∞ —
GLM-4.6 Z.ai integer — 1…+∞ —
GLM-4.6 Subscription Z.ai integer — 1…+∞ —
GLM-4.7 Z.ai integer — 1…+∞ —
GLM-4.7 Subscription Z.ai integer — 1…+∞ —
GLM-4.7-Flash Z.ai integer — 1…+∞ —
GLM-4.7-FlashX Z.ai integer — 1…+∞ —
GLM-5 Z.ai integer — 1…+∞ —
GLM-5 Subscription Z.ai integer — 1…+∞ —
GLM-5-Turbo Z.ai integer — 1…+∞ —
GLM-5-Turbo Subscription Z.ai integer — 1…+∞ —
GLM-5.1 Z.ai integer — 1…+∞ —
GLM-5.1 Subscription Z.ai integer — 1…+∞ —

Models without max_tokens documented yet

These models don't list max_tokens in the catalog. That may mean the provider doesn't accept it, or it simply hasn't been added yet — the data is community-maintained.

Google

Gemini 2.5 Flash Gemini 2.5 Flash Gemini 2.5 Flash Lite Gemini 2.5 Flash Lite Gemini 2.5 Pro Gemini 2.5 Pro Gemini 3 Flash Preview Gemini 3.1 Flash Lite Gemini 3.1 Flash Lite Preview Gemini 3.1 Pro Preview Gemini 3.5 Flash

Meta

Llama 3.3 70B Instruct Llama 3.3 8B Instruct Llama 4 Maverick 17B 128E Instruct FP8 Llama 4 Scout 17B 16E Instruct FP8

MiniMax

MiniMax M2 MiniMax M2.1 MiniMax M2.1 Highspeed MiniMax M2.5 MiniMax M2.5 Highspeed MiniMax M2.7 MiniMax M2.7 Highspeed Minimax M3

Moonshot AI

Kimi K2.5 Kimi K2.6 Kimi K2.6 Kimi K2.7 Code Kimi K2.7 Code Highspeed Moonshot v1 128K Moonshot v1 32K Moonshot v1 8K

Nvidia

Gliner Pii Llama 3.1 Nemotron Safety Guard 8b V3 Nemoguard Jailbreak Detect

OpenAI

Gpt 5 Gpt 5 Chat Latest Gpt 5 Mini Gpt 5 Nano Gpt 5.1 Gpt 5.1 Codex Gpt 5.1 Codex Max Gpt 5.2 Gpt 5.2 Gpt 5.2 Codex Gpt 5.3 Codex Gpt 5.3 Codex Gpt 5.3 Codex Spark Gpt 5.4 Gpt 5.4 Gpt 5.4 Mini Gpt 5.4 Mini Gpt 5.4 Nano Gpt 5.4 Pro Gpt 5.4 Pro Gpt 5.5 Gpt 5.5 Gpt 5.5 Pro Gpt 5.5 Pro o1 o3 O3 Pro o3-mini o4-mini

xAI

Grok 4.20 0309 Non Reasoning Grok 4.20 0309 Reasoning Grok 4.20 Multi Agent 0309 Grok 4.3 Grok Build 0.1

Xiaomi

Mimo V2.5 Mimo V2.5 Mimo V2.5 Pro

Related length parameters

Max completion tokens max_completion_tokens Stop sequence stop Max output tokens generationConfig.maxOutputTokens Stop sequences stop_sequences Number of completions n Max output tokens max_output_tokens

Resources

All parameters Use the API Full catalog

How to use

Building with an AI agent? Hit Copy to grab this whole guide as Markdown and paste it in — or point your agent straight at /llms.txt.

modelparams.dev is an open, community-maintained catalog of model parameters. Each entry shows the knobs you can turn — type, default, range, and the conditions that gate it.

The same model accessed via an API key and via a subscription usually exposes a different set of parameters. We list both as separate entries so the data stays honest.

Catalog API

The full catalog is static JSON, CORS-enabled, served from the edge.

curl https://modelparams.dev/api/v1/models.json

Each entry is keyed by provider/model for API-key variants; subscription variants append -subscription.

If you only need the params for one model contract, use the providerless endpoint. Subscription contracts are model slugs with -subscription. 

curl https://modelparams.dev/api/v1/params/gpt-5.5.json
curl https://modelparams.dev/api/v1/params/gpt-5.5-subscription.json

Single model

curl https://modelparams.dev/api/v1/models/anthropic/claude-opus-4-7.json
curl https://modelparams.dev/api/v1/models/anthropic/claude-opus-4-7-subscription.json

JSON Schema

Every entry validates against a JSON Schema you can use in your editor or pipeline. 

curl https://modelparams.dev/api/v1/schema.json

Add this header to any YAML you author for autocomplete in VS Code: 

# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json

Logos

Provider logos are available at /assets/logos/{provider}.svg where {provider} is the provider slug. They use currentColor so they inherit your text color. 

curl https://modelparams.dev/assets/logos/anthropic.svg

Logos are sourced from the models.dev repo (MIT) and used under nominative fair use.

Contribute

The data lives in YAML under models/{provider}/{model}-{auth}.yaml in the GitHub repo. Open a PR; CI validates against the schema and rebuilds.

Edit on GitHub MIT licensed