AI
Models
Last updated April 17, 2026
Browse all supported AI models available through Cencori, including their providers, types, and context windows.
Overview
Cencori provides unified access to 100+ state-of-the-art AI models from leading providers like OpenAI, Anthropic, Google, Mistral, xAI, DeepSeek, and more.
You can use any of these models with the same API format, simply by changing the model parameter.
Model Catalog
Claude 3.5 Haiku
claude-3-5-haiku-20241022
Fast and efficient
Claude 3.5 Sonnet
claude-3-5-sonnet-20241022
Balance of speed and capability
Claude 3.7 Sonnet
claude-3-7-sonnet
Hybrid reasoning model
Claude Haiku 4.5
claude-haiku-4.5
Fastest Claude model
Claude Opus 4
claude-opus-4
Most capable Claude model
Claude Opus 4.5
claude-opus-4.5
Latest, most intelligent
Claude Opus 4.6
claude-opus-4.6
Latest flagship, agentic coding record-breaker
Claude Opus 4.7
claude-opus-4.7
Latest flagship, improved reasoning & agentic coding
Claude Sonnet 4
claude-sonnet-4
Balanced speed & capability
Claude Sonnet 4.5
claude-sonnet-4.5
Enhanced coding & agents
Claude Sonnet 4.6
claude-sonnet-4.6
Latest flagship, enhanced reasoning & coding
Command A
command-a-03-2025
Most performant, agentic tasks
Command Light
command-light
Fast and efficient
Command R
command-r
Balanced performance
Command R+
command-r-plus-08-2024
Complex RAG and multi-step
DeepSeek Coder V2
deepseek-coder-v2
338 languages, GPT-4 level
DeepSeek R1
deepseek-reasoner
Reasoning model
DeepSeek V3
deepseek-chat
128K context, MIT license
DeepSeek V3.1
deepseek-v3.1
Hybrid thinking modes
DeepSeek V3.2
deepseek-v3.2
GPT-5 level, daily driver
DeepSeek V3.2 Speciale
deepseek-v3.2-speciale
Maxed reasoning, competition gold
Gemini 2.0 Flash
gemini-2.0-flash
Fast model
Gemini 2.0 Flash Thinking
gemini-2.0-flash-thinking
Reasoning variant
Gemini 2.5 Flash
gemini-2.5-flash
Thinking capabilities
Gemini 2.5 Flash Lite
gemini-2.5-flash-lite
Speed optimized
Gemini 2.5 Pro
gemini-2.5-pro
Enhanced reasoning & coding
Gemini 3 Deep Think
gemini-3-deep-think
Deep iterative reasoning
Gemini 3 Flash
gemini-3-flash
Frontier speed & intelligence
Gemini 3 Pro
gemini-3-pro
Powerful Gemini model
Gemini 3 Pro Image
gemini-3-pro-image
Fast photorealism
Gemini 3.1 Flash Image (Nano Banana 2)
gemini-3.1-flash-image
Reasoning-guided image synthesis, up to 4K
Gemini 3.1 Pro (Custom Tools)
gemini-3.1-pro-preview-customtools
Optimized for custom tools and bash
Gemini 3.1 Pro Preview
gemini-3.1-pro-preview
Latest flagship preview, 1M context, enhanced reasoning
Imagen 3
imagen-3
High quality images
Llama 3.1 8B Instant
llama-3.1-8b-instant
Ultra-fast inference
Llama 3.3 70B Versatile
llama-3.3-70b-versatile
Groq-hosted versatile Llama 3.3 model
Llama 4 Maverick
llama-4-maverick
Latest multimodal Llama
Llama 4 Scout
llama-4-scout
Advanced Llama 4 model
Mixtral 8x7B
mixtral-8x7b-32768
MoE architecture
Llama 3.3 70B
meta-llama/Llama-3.3-70B-Instruct
Via HF Inference
Llama 4 Maverick
meta-llama/Llama-4-Maverick
Via HF Inference
Mistral Large 3
mistralai/Mistral-Large-3
Via HF Inference
Qwen 2.5 72B
Qwen/Qwen2.5-72B-Instruct
Via HF Inference
Llama 3.1 405B
llama-3.1-405b
Largest open model
Llama 3.1 70B
llama-3.1-70b
Balanced performance
Llama 3.2 90B Vision
llama-3.2-90b-vision
Multimodal understanding
Llama 3.3 70B
llama-3.3-70b
Latest Llama 3 model
Llama 4 Maverick
llama-4-maverick
Latest multimodal flagship
Llama 4 Scout
llama-4-scout
Advanced reasoning
Codestral 25.01
codestral-latest
2.5x faster code generation
Devstral 2
devstral-latest
Frontier code agents
Magistral Medium
magistral-medium
Multimodal reasoning
Ministral 3B
ministral-3b
Compact edge model
Ministral 8B
ministral-8b
Small efficient model
Mistral Large 3
mistral-large-latest
675B params, best open-weight multimodal
Mistral Medium 3.1
mistral-medium-latest
Frontier-class multimodal
Mistral Small 3
mistral-small-latest
24B params, fast
DALL-E 2
dall-e-2
Fast image generation
DALL-E 3
dall-e-3
High quality images
GPT Image 1
gpt-image-1
ChatGPT image generation model
GPT Image 1.5
gpt-image-1.5
Best text rendering
GPT-4 Turbo
gpt-4-turbo
Legacy GPT-4 model
GPT-4.1
gpt-4.1
Long-context GPT-4.1
GPT-4.1 Mini
gpt-4.1-mini
Balanced GPT-4.1 model
GPT-4.1 Nano
gpt-4.1-nano
Fast GPT-4.1 nano model
GPT-4o
gpt-4o
Omni-modal model
GPT-4o Mini
gpt-4o-mini
Fast and cost-effective
GPT-5
gpt-5
Flagship model
GPT-5 Mini
gpt-5-mini
Fast and efficient
GPT-5 Nano
gpt-5-nano
Lowest-latency GPT-5 model
GPT-5 Pro
gpt-5-pro
High-quality GPT-5 variant
GPT-5.1
gpt-5.1
Improved GPT-5 generation
GPT-5.2
gpt-5.2
Latest GPT-5.2 flagship
GPT-5.2 Pro
gpt-5.2-pro
Most capable GPT-5.2 variant
GPT-5.3 Instant
gpt-5.3-chat-latest
Latest GPT-5.3 instant release
GPT-5.4 Pro
gpt-5.4-pro
Most capable GPT-5.4 variant
GPT-5.4 Thinking
gpt-5.4
Latest GPT-5.4 reasoning model
o1
o1
Legacy reasoning model
o3
o3
Advanced reasoning model
o3 Mini
o3-mini
Fast reasoning model
o3 Pro
o3-pro
Most advanced reasoning model
o4 Mini
o4-mini
Successor to o1-mini
Claude Opus 4.5 (via OpenRouter)
anthropic/claude-opus-4.5
Unified billing
Gemini 3 Pro (via OpenRouter)
google/gemini-3-pro
Meta-provider
GPT-5 (via OpenRouter)
openai/gpt-5
Access any model
Grok 4 (via OpenRouter)
x-ai/grok-4
Access xAI models
Sonar
sonar
Default web-connected
Sonar Large Online
llama-3.1-sonar-large-128k-online
Web-connected search
Sonar Pro
sonar-pro
Enhanced search, richer context
Sonar Reasoning Pro
sonar-reasoning-pro
Deep inference & research
Qwen 2.5 32B
qwen2.5-32b-instruct
Balanced performance
Qwen 2.5 72B
qwen2.5-72b-instruct
Flagship model
Qwen 2.5 Coder 32B
qwen2.5-coder-32b
Code specialized
QwQ 32B
qwq-32b-preview
Reasoning model
DeepSeek V3.1
deepseek-ai/DeepSeek-V3.1
Hybrid reasoning
Llama 3.3 70B Turbo
meta-llama/Llama-3.3-70B-Instruct-Turbo
Fast Llama inference
Llama 4 Maverick
meta-llama/Llama-4-Maverick
Latest Llama
Qwen 2.5 72B
Qwen/Qwen2.5-72B-Instruct-Turbo
Alibaba flagship
Grok 3
grok-3
DeepSearch, Big Brain Mode
Grok 3 Mini
grok-3-mini
Fast responses
Grok 4
grok-4
Enhanced reasoning, real-time search
Grok 4 Heavy
grok-4-heavy
Maximum capability
Grok 4.1
grok-4.1
Improved multimodal & reasoning
Grok 4.1 Fast
grok-4.1-fast
Best agentic tool calling
Grok Code Fast
grok-code-fast-1
Fast agentic coding
Model | Provider | Type | Context | Description | Status |
|---|---|---|---|---|---|
Claude 3.5 Haiku claude-3-5-haiku-20241022 | Anthropic | chat | 200K | Fast and efficient | Live |
Claude 3.5 Sonnet claude-3-5-sonnet-20241022 | Anthropic | chat | 200K | Balance of speed and capability | Live |
Claude 3.7 Sonnet claude-3-7-sonnet | Anthropic | reasoning | 200K | Hybrid reasoning model | Live |
Claude Haiku 4.5 claude-haiku-4.5 | Anthropic | chat | 200K | Fastest Claude model | Live |
Claude Opus 4 claude-opus-4 | Anthropic | chat | 200K | Most capable Claude model | Live |
Claude Opus 4.5 claude-opus-4.5 | Anthropic | chat | 200K | Latest, most intelligent | Live |
Claude Opus 4.6 claude-opus-4.6 | Anthropic | chat | 1M | Latest flagship, agentic coding record-breaker | Live |
Claude Opus 4.7 claude-opus-4.7 | Anthropic | chat | 1M | Latest flagship, improved reasoning & agentic coding | Live |
Claude Sonnet 4 claude-sonnet-4 | Anthropic | chat | 200K | Balanced speed & capability | Live |
Claude Sonnet 4.5 claude-sonnet-4.5 | Anthropic | chat | 200K | Enhanced coding & agents | Live |
Claude Sonnet 4.6 claude-sonnet-4.6 | Anthropic | chat | 200K | Latest flagship, enhanced reasoning & coding | Live |
Command A command-a-03-2025 | Cohere | chat | 256K | Most performant, agentic tasks | Live |
Command Light command-light | Cohere | chat | 4K | Fast and efficient | Live |
Command R command-r | Cohere | chat | 128K | Balanced performance | Live |
Command R+ command-r-plus-08-2024 | Cohere | chat | 128K | Complex RAG and multi-step | Live |
DeepSeek Coder V2 deepseek-coder-v2 | DeepSeek | code | 128K | 338 languages, GPT-4 level | Live |
DeepSeek R1 deepseek-reasoner | DeepSeek | reasoning | 64K | Reasoning model | Live |
DeepSeek V3 deepseek-chat | DeepSeek | chat | 128K | 128K context, MIT license | Live |
DeepSeek V3.1 deepseek-v3.1 | DeepSeek | chat | 128K | Hybrid thinking modes | Live |
DeepSeek V3.2 deepseek-v3.2 | DeepSeek | chat | 128K | GPT-5 level, daily driver | Live |
DeepSeek V3.2 Speciale deepseek-v3.2-speciale | DeepSeek | reasoning | 128K | Maxed reasoning, competition gold | Live |
Gemini 2.0 Flash gemini-2.0-flash | Google AI | chat | 1M | Fast model | Live |
Gemini 2.0 Flash Thinking gemini-2.0-flash-thinking | Google AI | reasoning | 1M | Reasoning variant | Live |
Gemini 2.5 Flash gemini-2.5-flash | Google AI | chat | 1M | Thinking capabilities | Live |
Gemini 2.5 Flash Lite gemini-2.5-flash-lite | Google AI | chat | 1M | Speed optimized | Live |
Gemini 2.5 Pro gemini-2.5-pro | Google AI | chat | 1M | Enhanced reasoning & coding | Live |
Gemini 3 Deep Think gemini-3-deep-think | Google AI | reasoning | 1M | Deep iterative reasoning | Live |
Gemini 3 Flash gemini-3-flash | Google AI | chat | 1M | Frontier speed & intelligence | Live |
Gemini 3 Pro gemini-3-pro | Google AI | chat | 2M | Powerful Gemini model | Live |
Gemini 3 Pro Image gemini-3-pro-image | Google AI | image | — | Fast photorealism | Live |
Gemini 3.1 Flash Image (Nano Banana 2) gemini-3.1-flash-image | Google AI | image | — | Reasoning-guided image synthesis, up to 4K | Live |
Gemini 3.1 Pro (Custom Tools) gemini-3.1-pro-preview-customtools | Google AI | chat | 1M | Optimized for custom tools and bash | Live |
Gemini 3.1 Pro Preview gemini-3.1-pro-preview | Google AI | chat | 1M | Latest flagship preview, 1M context, enhanced reasoning | Live |
Imagen 3 imagen-3 | Google AI | image | — | High quality images | Live |
Llama 3.1 8B Instant llama-3.1-8b-instant | Groq | chat | 128K | Ultra-fast inference | Live |
Llama 3.3 70B Versatile llama-3.3-70b-versatile | Groq | chat | 128K | Groq-hosted versatile Llama 3.3 model | Live |
Llama 4 Maverick llama-4-maverick | Groq | chat | 256K | Latest multimodal Llama | Live |
Llama 4 Scout llama-4-scout | Groq | chat | 256K | Advanced Llama 4 model | Live |
Mixtral 8x7B mixtral-8x7b-32768 | Groq | chat | 33K | MoE architecture | Live |
Llama 3.3 70B meta-llama/Llama-3.3-70B-Instruct | Hugging Face | chat | 128K | Via HF Inference | Live |
Llama 4 Maverick meta-llama/Llama-4-Maverick | Hugging Face | chat | 256K | Via HF Inference | Live |
Mistral Large 3 mistralai/Mistral-Large-3 | Hugging Face | chat | 128K | Via HF Inference | Live |
Qwen 2.5 72B Qwen/Qwen2.5-72B-Instruct | Hugging Face | chat | 32K | Via HF Inference | Live |
Llama 3.1 405B llama-3.1-405b | Meta AI | chat | 128K | Largest open model | Live |
Llama 3.1 70B llama-3.1-70b | Meta AI | chat | 128K | Balanced performance | Live |
Llama 3.2 90B Vision llama-3.2-90b-vision | Meta AI | chat | 128K | Multimodal understanding | Live |
Llama 3.3 70B llama-3.3-70b | Meta AI | chat | 128K | Latest Llama 3 model | Live |
Llama 4 Maverick llama-4-maverick | Meta AI | chat | 256K | Latest multimodal flagship | Live |
Llama 4 Scout llama-4-scout | Meta AI | chat | 256K | Advanced reasoning | Live |
Codestral 25.01 codestral-latest | Mistral AI | code | 256K | 2.5x faster code generation | Live |
Devstral 2 devstral-latest | Mistral AI | code | 256K | Frontier code agents | Live |
Magistral Medium magistral-medium | Mistral AI | reasoning | 128K | Multimodal reasoning | Live |
Ministral 3B ministral-3b | Mistral AI | chat | 128K | Compact edge model | Live |
Ministral 8B ministral-8b | Mistral AI | chat | 128K | Small efficient model | Live |
Mistral Large 3 mistral-large-latest | Mistral AI | chat | 128K | 675B params, best open-weight multimodal | Live |
Mistral Medium 3.1 mistral-medium-latest | Mistral AI | chat | 128K | Frontier-class multimodal | Live |
Mistral Small 3 mistral-small-latest | Mistral AI | chat | 32K | 24B params, fast | Live |
DALL-E 2 dall-e-2 | OpenAI | image | — | Fast image generation | Live |
DALL-E 3 dall-e-3 | OpenAI | image | — | High quality images | Live |
GPT Image 1 gpt-image-1 | OpenAI | image | — | ChatGPT image generation model | Live |
GPT Image 1.5 gpt-image-1.5 | OpenAI | image | — | Best text rendering | Live |
GPT-4 Turbo gpt-4-turbo | OpenAI | chat | 128K | Legacy GPT-4 model | Live |
GPT-4.1 gpt-4.1 | OpenAI | code | 1.0M | Long-context GPT-4.1 | Live |
GPT-4.1 Mini gpt-4.1-mini | OpenAI | chat | 1.0M | Balanced GPT-4.1 model | Live |
GPT-4.1 Nano gpt-4.1-nano | OpenAI | chat | 1.0M | Fast GPT-4.1 nano model | Live |
GPT-4o gpt-4o | OpenAI | chat | 128K | Omni-modal model | Live |
GPT-4o Mini gpt-4o-mini | OpenAI | chat | 128K | Fast and cost-effective | Live |
GPT-5 gpt-5 | OpenAI | chat | 400K | Flagship model | Live |
GPT-5 Mini gpt-5-mini | OpenAI | chat | 400K | Fast and efficient | Live |
GPT-5 Nano gpt-5-nano | OpenAI | chat | 400K | Lowest-latency GPT-5 model | Live |
GPT-5 Pro gpt-5-pro | OpenAI | chat | 400K | High-quality GPT-5 variant | Live |
GPT-5.1 gpt-5.1 | OpenAI | chat | 400K | Improved GPT-5 generation | Live |
GPT-5.2 gpt-5.2 | OpenAI | chat | 400K | Latest GPT-5.2 flagship | Live |
GPT-5.2 Pro gpt-5.2-pro | OpenAI | chat | 400K | Most capable GPT-5.2 variant | Live |
GPT-5.3 Instant gpt-5.3-chat-latest | OpenAI | chat | 400K | Latest GPT-5.3 instant release | Live |
GPT-5.4 Pro gpt-5.4-pro | OpenAI | chat | 400K | Most capable GPT-5.4 variant | Live |
GPT-5.4 Thinking gpt-5.4 | OpenAI | chat | 400K | Latest GPT-5.4 reasoning model | Live |
o1 o1 | OpenAI | reasoning | 200K | Legacy reasoning model | Live |
o3 o3 | OpenAI | reasoning | 200K | Advanced reasoning model | Live |
o3 Mini o3-mini | OpenAI | reasoning | 200K | Fast reasoning model | Live |
o3 Pro o3-pro | OpenAI | reasoning | 200K | Most advanced reasoning model | Live |
o4 Mini o4-mini | OpenAI | reasoning | 200K | Successor to o1-mini | Live |
Claude Opus 4.5 (via OpenRouter) anthropic/claude-opus-4.5 | OpenRouter | chat | 200K | Unified billing | Live |
Gemini 3 Pro (via OpenRouter) google/gemini-3-pro | OpenRouter | chat | 2M | Meta-provider | Live |
GPT-5 (via OpenRouter) openai/gpt-5 | OpenRouter | chat | 256K | Access any model | Live |
Grok 4 (via OpenRouter) x-ai/grok-4 | OpenRouter | chat | 256K | Access xAI models | Live |
Sonar sonar | Perplexity | search | 128K | Default web-connected | Live |
Sonar Large Online llama-3.1-sonar-large-128k-online | Perplexity | search | 128K | Web-connected search | Live |
Sonar Pro sonar-pro | Perplexity | search | 128K | Enhanced search, richer context | Live |
Sonar Reasoning Pro sonar-reasoning-pro | Perplexity | reasoning | 128K | Deep inference & research | Live |
Qwen 2.5 32B qwen2.5-32b-instruct | Qwen | chat | 128K | Balanced performance | Live |
Qwen 2.5 72B qwen2.5-72b-instruct | Qwen | chat | 128K | Flagship model | Live |
Qwen 2.5 Coder 32B qwen2.5-coder-32b | Qwen | code | 128K | Code specialized | Live |
QwQ 32B qwq-32b-preview | Qwen | reasoning | 32K | Reasoning model | Live |
DeepSeek V3.1 deepseek-ai/DeepSeek-V3.1 | Together AI | chat | 128K | Hybrid reasoning | Live |
Llama 3.3 70B Turbo meta-llama/Llama-3.3-70B-Instruct-Turbo | Together AI | chat | 128K | Fast Llama inference | Live |
Llama 4 Maverick meta-llama/Llama-4-Maverick | Together AI | chat | 256K | Latest Llama | Live |
Qwen 2.5 72B Qwen/Qwen2.5-72B-Instruct-Turbo | Together AI | chat | 32K | Alibaba flagship | Live |
Grok 3 grok-3 | xAI | chat | 128K | DeepSearch, Big Brain Mode | Live |
Grok 3 Mini grok-3-mini | xAI | chat | 128K | Fast responses | Live |
Grok 4 grok-4 | xAI | chat | 256K | Enhanced reasoning, real-time search | Live |
Grok 4 Heavy grok-4-heavy | xAI | chat | 256K | Maximum capability | Live |
Grok 4.1 grok-4.1 | xAI | chat | 256K | Improved multimodal & reasoning | Live |
Grok 4.1 Fast grok-4.1-fast | xAI | chat | 2M | Best agentic tool calling | Live |
Grok Code Fast grok-code-fast-1 | xAI | code | 128K | Fast agentic coding | Live |
Usage
To use a model, pass its ID to any of the AI SDK methods:
import { cencori } from '@/lib/cencori'
const result = await cencori.ai.chat({
model: 'gpt-4o', // Use the ID from the catalog above
messages: [
{ role: 'user', content: 'Hello!' }
]
})