INDEPENDENT MODEL EXPLORER

Your next model.
Start with evidence.

See the capability, the cost, and the gaps. Build a shortlist worth testing against your own work.

Recent snapshotData snapshot cf-1.3

01 / DISCOVER

A starting point for your work.

Explore the directory →
01Coding

Claude Fable 5

A high-scoring candidate in this measured dimension. Check the gaps and test it on your workload.

View evidence ↗
02Agent workflows

Claude Opus 5.5

A high-scoring candidate in this measured dimension. Check the gaps and test it on your workload.

View evidence ↗
03Reasoning

GPT-6 Astra

A high-scoring candidate in this measured dimension. Check the gaps and test it on your workload.

View evidence ↗

02 / THE FRONTIER

More than a position on a chart.

399 Models · cf-1.3
Anthropic

Claude Opus 5.5

MathematicsMultimodal
Cheese Index80.3Cheese Index
Input / output$4 / $20USD / 1M tokens
6 / 13Measured dimensions
Anthropic

Claude Fable 5

Mathematicsswe
Cheese Index76.4Cheese Index
Input / output$10 / $50USD / 1M tokens
9 / 13Measured dimensions
Anthropic

Claude Opus 5

MultimodalMathematics
Cheese Index75.6Cheese Index
Input / output$5 / $25USD / 1M tokens
7 / 13Measured dimensions
OpenAI

GPT-6.1 Sol

MathematicsReasoning
Cheese Index74.4Cheese Index
Input / output$2 / $10USD / 1M tokens
6 / 13Measured dimensions
Google

Gemini 4 Argon

ReasoningKnowledge
Cheese Index74.1Cheese Index
Input / output$2 / $10USD / 1M tokens
4 / 13Measured dimensions

Listed dataset prices; provider terms and dates need verification. Zero without explicit free-price evidence is treated as unknown.

03 / THINK BEFORE YOU SHIP

A good decision has a paper trail.

Trace the method, inspect missing evidence, and bring your own evaluation checklist.