Model catalog reference and public API availability.
This reference summarizes model identifiers in the configured catalog. Availability depends on the current deployment and your plan. Call GET /v1/models to discover the models available to your API key.
Catalog membership does not mean every model can be used with every endpoint. The public API documents chat completions, responses, and model discovery; image, video, speech, and embedding models below describe the wider catalog and are not a promise of corresponding public generation endpoints. Follow the API overview and each feature’s documentation.
OpenAI Models
GPT-5 Series
Model
Context
Best For
gpt-5.5
128K
Current flagship
gpt-5.4
128K
Mid-tier flagship
gpt-5.2
128K
Previous flagship
gpt-5.1
128K
Budget tier
gpt-5-mini
128K
Balanced, cost-effective
gpt-5-nano
128K
Fast, budget-friendly
GPT-4.1 Series
Model
Context
Best For
gpt-4.1
1M
Long context, vision
gpt-4.1-mini
1M
Efficient, vision
gpt-4.1-nano
1M
Ultra-fast, cheap
GPT-4 Series
Model
Context
Best For
gpt-4o
128K
General, Vision
gpt-4o-mini
128K
Cost-effective
Embeddings (catalog only)
The public API does not expose /v1/embeddings. These models are listed for catalog context, not as a callable public embedding service.
Model
Dimensions
Max Input
text-embedding-3-large
3072
8191
text-embedding-3-small
1536
8191
Anthropic Models
Claude 4.8 Series (Latest)
Model
Context
Best For
claude-opus-4.8
200K
Current flagship, agentic coding
Claude 4.7 Series
Model
Context
Best For
claude-opus-4.7
200K
Complex agentic tasks
Claude 4.6 Series
Model
Context
Best For
claude-sonnet-4.6
200K
Coding, Agents
claude-opus-4.6
200K
Most capable
Claude 4.5 Series
Model
Context
Best For
claude-sonnet-4.5
200K
Coding, Agents
claude-opus-4.5
200K
Most capable
claude-haiku-4.5
200K
Fast, affordable
Google Models
Gemini 3.5 Series (Latest)
Model
Context
Best For
gemini-3.5-flash
1M
Fast, beats 3.1 Pro on coding
Gemini 3.1 Series
Model
Context
Best For
gemini-3.1-pro-preview
1M
Most capable, reasoning
Gemini 3 Series
Model
Context
Best For
gemini-3-pro-preview
1M
Advanced reasoning
gemini-3-flash
1M
Fast, efficient
xAI Models
Grok 4.3 Series
Model
Context
Best For
grok-4.3
1M
Cost-efficient reasoning
Grok 4 Series
Model
Context
Best For
grok-4.6
1M
High-capability reasoning
grok-4.5
1M
Previous-generation reasoning
grok-4.20
1M
Agentic, multi-step work
grok-4.20-multi-agent
1M
Parallel multi-agent runs
grok-build-0.1
1M
Code generation
Mistral AI Models
Model
Context
Best For
mistral-large-3
256K
Most capable
mistral-medium-3.1
128K
Balanced
mistral-small-4
256K
Multimodal, agentic coding
mistral-small-3.2
128K
Fast, efficient
codestral-22b
256K
Code generation
DeepSeek Models
Model
Context
Best For
deepseek-v4-pro
128K
Flagship reasoning and coding
deepseek-v4-flash
128K
Fast, cost-effective V4
deepseek-v3.2
128K
General purpose
deepseek-v3.1
128K
Fast, versatile
Cohere Models
Model
Context
Best For
command-a
256K
Enterprise RAG
Perplexity Models
Model
Context
Best For
sonar-deep-research
128K
Deep research with citations
sonar-reasoning-pro
128K
Reasoning with web search
perplexity-sonar-large
128K
General search
perplexity-sonar
128K
Fast search
Meta Models
Llama 4 Series
Model
Context
Best For
llama4-scout
320K
Efficient, vision
llama4-maverick
1M
Complex tasks, vision
Llama 3.3 Series
Model
Context
Best For
llama-3.3-70b
128K
High quality
Open-Weight Models (GLM, Kimi, MiniMax, Qwen)
GLM Series (Zhipu AI)
Model
Context
Best For
glm-5.1
202K
Coding, agentic workflows
glm-5
202K
General purpose
glm-5v-turbo
203K
Native multimodal (text + image)
glm-4.7
128K
Balanced
glm-4.6
202K
Cost-efficient
Kimi Series (Moonshot AI)
Model
Context
Best For
kimi-k2.6
256K
Multimodal, long-horizon coding
kimi-k2
128K
Agentic workflows
kimi-k2-thinking
128K
Extended reasoning
MiniMax Series
Model
Context
Best For
minimax-m3
1M
Multimodal flagship
minimax-m2.7
200K
Balanced MoE
Qwen Series (Alibaba)
Model
Context
Best For
qwen3.7-max
1M
Long-horizon agent tasks
qwen3-coder-next
256K
Code generation
qwen3-32b
40K
Mid-size, versatile
NVIDIA Nemotron Series
Model
Context
Best For
nemotron-3-super-120b
1M
Flagship agentic reasoning
nemotron-3-nano-30b
256K
Fast, low-cost reasoning
Capability-Specific Models
Agentic Coding
Model
Context
Best For
grok-build-0.1
256K
xAI coding model (public beta)
gpt-5.3-codex
128K
OpenAI Responses API coding
Image Generation
Several previously listed image deployments are currently disabled and have been removed from this list. Check Studio for the generation options available to your account.
Model
Best For
gpt-image-2
Next-gen OpenAI image generation
gpt-image-1.5
Fast OpenAI image generation
Video Generation
Model
Best For
sora-2
OpenAI Sora 2
sora-2-pro
OpenAI Sora 2 Pro
The Veo models (veo-2, veo-3, veo-3-fast, veo-3.1, veo-3.1-fast) are no
longer available. Google retired Veo 2.0 and 3.0, and the Veo 3.1 tiers are not
currently served. Sora is the video generation route.
Audio
Model
Best For
tts-1-hd
OpenAI HD text-to-speech
tts-1
OpenAI standard text-to-speech
gpt-4o-mini-tts
OpenAI mini text-to-speech
gpt-transcribe
OpenAI transcription (current)
gpt-live-transcribe
Low-latency streaming transcription
gpt-4o-transcribe
GPT-4o powered transcription
gpt-4o-mini-transcribe
Cheaper GPT-4o transcription
whisper-1
OpenAI Whisper transcription
whisper-1, gpt-4o-transcribe and gpt-4o-mini-transcribe are scheduled for
retirement on 2027-02-26. gpt-transcribe (file transcription) and
gpt-live-transcribe (realtime streaming) are the replacements.
Search
Model
Best For
sonar-deep-research
Deep multi-step research
sonar-reasoning-pro
Reasoning with real-time sources
perplexity-sonar-large
Web-grounded answers
perplexity-sonar
Fast web search
Model Selection Guide
By Task
Task
Recommended
General chat
gpt-5-mini, claude-haiku-4.5
Complex reasoning
gpt-5.5, claude-opus-4.8, deepseek-v4-pro
Code generation
claude-sonnet-4.6, grok-build-0.1
Long documents
gpt-4.1, gemini-3.1-pro-preview
Web search
sonar-deep-research, perplexity-sonar-large
Embeddings (no public endpoint)
See the catalog-only note above
Fast inference
gpt-5-nano, gemini-3.1-flash-lite
Image generation (catalog)
gpt-image-2, gpt-image-1.5
Video generation
sora-2, sora-2-pro
Text-to-speech (catalog)
tts-1-hd, tts-1
Using Model IDs
Use the full model ID in API requests:
response = client.chat.completions.create( model="gpt-4o-mini", # Model ID messages=[...])
Use the ID exactly as listed above — the gateway matches it literally. Two
things that look reasonable but fail:
A provider prefix is not accepted.anthropic/claude-sonnet-4.5 returns a
model-not-found; the ID is claude-sonnet-4.5. (The only IDs containing a
slash are the self-hosted ollama/* models, where the slash is part of the
name itself.)
Versions use dots, not dashes.claude-sonnet-4.6 works;
claude-sonnet-4-6 does not.
Model Availability
Model availability may vary. Check the Model Hub for current availability and pricing.
