Guided model chooser

Find a model for the work

Apply practical constraints, then inspect a deterministic shortlist built from the same Index and coverage rules as the leaderboard.

Recommendations

Most capable

Claude Opus 5

Anthropic

62.2Overall Index

Standing: Rank 1 of 62
Leader gap: Leads shortlist
Coverage: 7 of 8 measured · 1 estimated
Access: Hosted API
Context: 1,000,000 tokens
Input / output: $5.0 / $25.0 per 1M

Official pricing (opens in a new tab)Checked Aug 3, 2026

Open model record

Lowest input price

DeepSeek V4 Flash

DeepSeek

50.0Overall Index

Standing: Rank 28 of 62
Leader gap: 12.1 Index points
Coverage: 8 of 8 measured · 0 estimated
Access: Open weights + hosted API
Context: 1,000,000 tokens
Input / output: $0.14 / $0.28 per 1M

Official pricing (opens in a new tab)Checked Aug 3, 2026

Open model record

Largest context

Llama 4 Scout

Meta

19.0Overall Index

Standing: Rank 62 of 62
Leader gap: 43.1 Index points
Coverage: 8 of 8 measured · 0 estimated
Access: Open weights; no first-party API price listed
Context: 10,000,000 tokens
Input / output: No first-party API price listed

Price source not listed

Open model record

Open-weights leader

GLM-5.2

Z.ai

56.3Overall Index

Standing: Rank 14 of 62
Leader gap: 5.9 Index points
Coverage: 8 of 8 measured · 0 estimated
Access: Open weights + hosted API
Context: 1,000,000 tokens
Input / output: $1.4 / $4.4 per 1M

Official pricing (opens in a new tab)Checked Aug 3, 2026

Open model record