AI Model Directory
Pricing & Specs

Every model you can benchmark on OpenMark — with live API pricing, context windows, and latency we measured ourselves. Updated whenever the registry changes.

102
models available to test
13
providers
$0.1–$600
output price range / 1M tokens

Prices below are per-token rates — not what your task will actually cost. A "cheap" model that needs 3x the tokens costs the same as a premium one. Click any model for full specs, or benchmark a shortlist on your own task to see real cost-per-task.

OpenAI (30 models)

ModelInput $/1MOutput $/1MContext
OpenAI o1-pro $150 $600 200K Specs →
GPT-5.4 Pro $30 $180 1.05M Specs →
GPT-5.5 Pro $30 $180 1.05M Specs →
GPT-5.2 pro $21 $168 400K Specs →
GPT-5 pro $15 $120 400K Specs →
OpenAI o3-pro $20 $80 200K Specs →
OpenAI o1 $15 $60 200K Specs →
GPT-5.5 $5 $30 1.05M Specs →
GPT-5.6 Sol $5 $30 1.05M Specs →
GPT-5.4 $2.5 $15 1.05M Specs →
GPT-5.2 $1.75 $14 400K Specs →
GPT-5.6 Terra $2 $12 1.05M Specs →
GPT-4o $2.5 $10 128K Specs →
GPT-5 $1.25 $10 400K Specs →
GPT-5.1 $1.25 $10 400K Specs →
GPT-4.1 $2 $8 1.04758M Specs →
OpenAI o3 $2 $8 200K Specs →
GPT-3.5 Turbo $3 $6 16.385K Specs →
Codex Mini $1.5 $6 200K Specs →
OpenAI o3-mini $1.1 $4.4 200K Specs →
OpenAI o4-mini $1.1 $4.4 200K Specs →
GPT-5.4 mini $0.75 $4.5 400K Specs →
GPT-5 Mini $0.25 $2 400K Specs →
GPT-4.1 Mini $0.4 $1.6 1.04758M Specs →
GPT-5.4 nano $0.2 $1.25 400K Specs →
GPT-5.6 Luna $0.2 $1.2 1.05M Specs →
GPT-4o mini $0.15 $0.6 128K Specs →
gpt-oss-120b $0.15 $0.6 131.072K Specs →
GPT-4.1 Nano $0.1 $0.4 1.04758M Specs →
GPT-5 Nano $0.05 $0.4 400K Specs →

Anthropic (15 models)

ModelInput $/1MOutput $/1MContext
Claude Opus 4 $15 $75 200K Specs →
Claude Opus 4.1 $15 $75 200K Specs →
Claude Fable 5 $10 $50 1M Specs →
Claude Opus 4.5 $5 $25 200K Specs →
Claude Opus 4.6 $5 $25 200K Specs →
Claude Opus 4.7 $5 $25 200K Specs →
Claude Opus 4.8 $5 $25 200K Specs →
Claude Opus 5 $5 $25 1M Specs →
Claude Sonnet 3.7 $3 $15 200K Specs →
Claude Sonnet 4 $3 $15 200K Specs →
Claude Sonnet 4.5 $3 $15 200K Specs →
Claude Sonnet 4.6 $3 $15 200K Specs →
Claude Sonnet 5 $3 $15 1M Specs →
Claude Haiku 4.5 $1 $5 200K Specs →
Claude Haiku 3 $0.25 $1.25 200K Specs →

Google (Gemini) (8 models)

ModelInput $/1MOutput $/1MContext
Gemini 3.1 Pro $2 $12 1.04858M Specs →
Gemini 2.5 Pro $1.25 $10 1.04858M Specs →
Gemini 3.5 Flash $1.5 $9 1.04858M Specs →
Gemini 3 Flash $0.5 $3 1.04858M Specs →
Gemini 2.5 Flash $0.3 $2.5 1.04858M Specs →
Gemini Robotics-ER 1.5 Preview $0.3 $2.5 1.04858M Specs →
Gemini 3.1 Flash-Lite $0.25 $1.5 1.04858M Specs →
Gemini 2.5 Flash-Lite $0.1 $0.4 1.04858M Specs →

xAI (12 models)

ModelInput $/1MOutput $/1MContext
Grok 3 $3 $15 131.072K Specs →
Grok 4 $3 $15 256K Specs →
Grok 4.5 $2 $6 500K Specs →
Grok 4.2 (Non-Reasoning) $1.25 $2.5 2M Specs →
Grok 4.2 (Reasoning) $1.25 $2.5 2M Specs →
Grok 4.3 $1.25 $2.5 2M Specs →
Grok Code Fast 1 $0.2 $1.5 256K Specs →
Grok 3 Mini $0.3 $0.5 131.072K Specs →
Grok 4 Fast (Non-Reasoning) $0.2 $0.5 2M Specs →
Grok 4 Fast $0.2 $0.5 2M Specs →
Grok 4.1 Fast (Non-Reasoning) $0.2 $0.5 2M Specs →
Grok 4.1 Fast $0.2 $0.5 2M Specs →

DeepSeek (3 models)

ModelInput $/1MOutput $/1MContext
DeepSeek-V4-Pro $1.32 $3.96 1M Specs →
DeepSeek-V4-Flash $0.44 $1.32 1M Specs →
DeepSeek-V4-Flash (Non-Reasoning) $0.44 $1.32 1M Specs →

Mistral (11 models)

ModelInput $/1MOutput $/1MContext
Mistral Medium 3.5 $1.5 $7.5 128K Specs →
Mixtral 8x22B $2 $6 65.536K Specs →
Magistral Small 1.1 $0.5 $1.5 128K Specs →
Mistral Large 3 $0.5 $1.5 256K Specs →
Mixtral 8x7B $0.7 $0.7 32.768K Specs →
Codestral 2508 $0.3 $0.9 128K Specs →
Mistral Small 4 $0.15 $0.6 256K Specs →
Mistral 7B $0.25 $0.25 32.768K Specs →
Ministral 14B $0.2 $0.2 256K Specs →
Ministral 8B $0.15 $0.15 256K Specs →
Ministral 3B $0.1 $0.1 256K Specs →

Meta (1 models)

ModelInput $/1MOutput $/1MContext
Llama 3.3 70B Instruct Turbo API $1.04 $1.04 131.072K Specs →

Qwen (Alibaba) (4 models)

ModelInput $/1MOutput $/1MContext
Qwen3.7 Max $1.25 $3.75 1M Specs →
Qwen 2.5 72B Instruct Turbo API $1.2 $1.2 131.072K Specs →
Qwen3-235B A22B fp8-tput $0.2 $0.6 40.96K Specs →
Qwen3.5 9B FP8 $0.17 $0.25 262.144K Specs →

MiniMax (9 models)

ModelInput $/1MOutput $/1MContext
MiniMax-M2.1-lightning $0.6 $4.8 196.608K Specs →
MiniMax-M2.5-lightning $0.6 $2.4 196.608K Specs →
MiniMax-M2.7-highspeed $0.6 $2.4 196.608K Specs →
MiniMax-M2 $0.3 $1.2 196.608K Specs →
MiniMax-M2-her $0.3 $1.2 32.768K Specs →
MiniMax-M2.1 $0.3 $1.2 196.608K Specs →
MiniMax-M2.5 $0.3 $1.2 196.608K Specs →
MiniMax-M2.7 $0.3 $1.2 196.608K Specs →
MiniMax-M3 $0.3 $1.2 1M Specs →

Zhipu AI (2 models)

ModelInput $/1MOutput $/1MContext
GLM-5.2 $1.4 $4.4 262.144K Specs →
Glm 4.6 Fp8 $0.6 $2.2 202.752K Specs →

Perplexity (3 models)

ModelInput $/1MOutput $/1MContext
Sonar Pro $3 $15 200K Specs →
Sonar Reasoning Pro $2 $8 128K Specs →
Sonar $1 $1 128K Specs →

Cohere (3 models)

ModelInput $/1MOutput $/1MContext
Command A $2.5 $10 256K Specs →
Command R $0.15 $0.6 128K Specs →
Command R7B $0.0375 $0.15 132K Specs →

Deep Cogito (1 models)

ModelInput $/1MOutput $/1MContext
Cogito v2.1 671B $1.25 $1.25 163.8K Specs →

Retired & superseded models

No longer offered for benchmarking, kept for reference. If you still run one of these in production, its successor is probably cheaper and better — worth a re-test.

Stop Reading Specs. Start Measuring.

Describe your task, pick models from this list, and get ranked results
with real accuracy, cost and latency — in minutes. Free tier available.

Benchmark Your Task — Free →