Skip to content

Model catalog / Preview

Find a model
for your workload.

Compare tasks, context windows and reference rates. This preview catalog is for exploration; models and commercial terms are confirmed individually for each pilot.

Filters
Provider⌄
Capability⌄
Billing type⌄

Reference rates, not a quote. Copy a model ID to use in the model field of an API request; copying does not enable access.

A

Claude Haiku 4.5

Anthropic

Pilot availability to confirm

Snappy, inexpensive model for classification and routing.

ChatTools
Context window
200K tokens
Input / 1M tokens
$0.80
Output / 1M tokens
$4.00
anthropic/claude-haiku-4.5
A

Claude Opus 4.8

Anthropic

Pilot availability to confirm

Deepest reasoning and long-horizon coding from Anthropic.

ChatToolsReasoningCodingLong context
Context window
500K tokens
Input / 1M tokens
$6.00
Output / 1M tokens
$30.00
anthropic/claude-opus-4.8
A

Claude Sonnet 4.6

Anthropic

Pilot availability to confirm

Balanced speed, cost, and quality for everyday agents.

ChatToolsCodingVision
Context window
400K tokens
Input / 1M tokens
$3.00
Output / 1M tokens
$15.00
anthropic/claude-sonnet-4.6
M

Codestral 2

Mistral

Pilot availability to confirm

Code-specialized model tuned for fill-in-the-middle and repos.

CodingTools
Context window
256K tokens
Input / 1M tokens
$0.90
Output / 1M tokens
$2.80
mistral/codestral-2
C

Command A

Cohere

Pilot availability to confirm

Enterprise RAG-optimized model with tool use.

ChatTools
Context window
256K tokens
Input / 1M tokens
$2.50
Output / 1M tokens
$10.00
cohere/command-a
D

DeepSeek R1

DeepSeek

Pilot availability to confirm

Open reasoning model with visible chain-of-thought.

ReasoningChat
Context window
128K tokens
Input / 1M tokens
$0.55
Output / 1M tokens
$2.20
deepseek/deepseek-r1
D

DeepSeek V3.2

DeepSeek

Pilot availability to confirm

Cost-efficient reasoning model popular for coding.

ChatCodingTools
Context window
128K tokens
Input / 1M tokens
$0.27
Output / 1M tokens
$1.10
deepseek/deepseek-v3.2
O

Embedding 3 Large

OpenAI

Pilot availability to confirm

High-quality text embeddings for search and RAG.

Embeddings
Context window
8K tokens
Input / 1M tokens
$0.13
Output / 1M tokens
$0.00
openai/text-embedding-3-large
B

FLUX 2 Pro

Black Forest

Pilot availability to confirm

State-of-the-art open image model with precise prompt adherence.

Drawing
Reference rate
$0.05 / image
bfl/flux-2-pro
G

Gemini 2.5 Flash

Google

Pilot availability to confirm

Ultra-cheap long context for retrieval-heavy pipelines.

ChatVisionLong context
Context window
1M tokens
Input / 1M tokens
$0.30
Output / 1M tokens
$1.20
google/gemini-2.5-flash
G

Gemini 2.5 Pro

Google

Pilot availability to confirm

Million-token context with strong multimodal grounding.

ChatToolsVisionReasoningLong context
Context window
1M tokens
Input / 1M tokens
$2.50
Output / 1M tokens
$10.00
google/gemini-2.5-pro
O

GPT Image 2

OpenAI

Pilot availability to confirm

Fast, instruction-following image generation and editing.

Drawing
Reference rate
$0.03 / image
openai/gpt-image-2
O

GPT-5

OpenAI

Pilot availability to confirm

Frontier reasoning and tool use across text and images.

ChatToolsReasoningVision
Context window
400K tokens
Input / 1M tokens
$5.00
Output / 1M tokens
$15.00
openai/gpt-5
O

GPT-5 Mini

OpenAI

Pilot availability to confirm

Fast, low-cost model for high-volume production workloads.

ChatToolsVision
Context window
200K tokens
Input / 1M tokens
$0.40
Output / 1M tokens
$1.60
openai/gpt-5-mini
O

GPT-5 Nano

OpenAI

Pilot availability to confirm

Smallest GPT-5 tier for classification, routing, and autocomplete.

ChatTools
Context window
128K tokens
Input / 1M tokens
$0.05
Output / 1M tokens
$0.40
openai/gpt-5-nano
X

Grok 4

xAI

Pilot availability to confirm

Real-time-aware model with strong reasoning.

ChatReasoningToolsVision
Context window
256K tokens
Input / 1M tokens
$3.00
Output / 1M tokens
$15.00
xai/grok-4
X

Grok 4 Fast

xAI

Pilot availability to confirm

Latency-optimized Grok tier for interactive products.

ChatTools
Context window
128K tokens
Input / 1M tokens
$0.50
Output / 1M tokens
$1.50
xai/grok-4-fast
G

Imagen 4

Google

Pilot availability to confirm

Photorealistic text-to-image generation.

Drawing
Reference rate
$0.04 / image
google/imagen-4
M

Kimi K2

Moonshot

Pilot availability to confirm

Open trillion-parameter MoE with excellent tool use.

ChatToolsCoding
Context window
256K tokens
Input / 1M tokens
$0.60
Output / 1M tokens
$2.40
moonshot/kimi-k2
M

Llama 4 Maverick

Meta

Pilot availability to confirm

Open-weight MoE model with strong general performance.

ChatToolsVision
Context window
256K tokens
Input / 1M tokens
$0.50
Output / 1M tokens
$1.50
meta/llama-4-maverick
M

Llama 4 Scout

Meta

Pilot availability to confirm

10M-token context window for whole-corpus analysis.

ChatLong context
Context window
10M tokens
Input / 1M tokens
$0.15
Output / 1M tokens
$0.50
meta/llama-4-scout
M

Mistral Large 3

Mistral

Pilot availability to confirm

European frontier model with strong multilingual support.

ChatTools
Context window
256K tokens
Input / 1M tokens
$2.00
Output / 1M tokens
$6.00
mistral/mistral-large-3
Q

Qwen3 Coder

Qwen

Pilot availability to confirm

Agentic coding model with strong repository-scale edits.

CodingTools
Context window
256K tokens
Input / 1M tokens
$0.90
Output / 1M tokens
$3.60
qwen/qwen3-coder
Q

Qwen3 Max

Qwen

Pilot availability to confirm

Alibaba's flagship multilingual model.

ChatVisionTools
Context window
256K tokens
Input / 1M tokens
$1.20
Output / 1M tokens
$6.00
qwen/qwen3-max
O

Sora 2

OpenAI

Pilot availability to confirm

Text- and image-conditioned video generation with audio.

Video
Reference rate
$0.10 / second
openai/sora-2
G

Veo 3.1

Google

Pilot availability to confirm

Cinematic video generation with native soundtrack.

Video
Reference rate
$0.15 / second
google/veo-3.1
O

Whisper 3

OpenAI

Pilot availability to confirm

Robust multilingual speech-to-text transcription.

Speech
Reference rate
$0.006 / minute
openai/whisper-3

Need help choosing? Tell us about your workload.

Discuss your use case