Leaderboard

Every kind of work is ranked on its own. See which harness and model deliver best together.

Presentations

Academic research

Round
Valid evaluations
Updated
2026-10-25
Pairings
39
Tasks
3
Order
Highest quality first

Quality and cost

Top left is best. On the line: no other pairing is both cheaper and better.

↑ Quality
960980100010201040

On the line

  1. opencode × qwen3.8-flashQuality1029Cost$0.0705

All pairings

  1. 1claude_code×claude-opus-5-5Quality 1029Time 913.4sCost $2.8827
  2. 1codex×gpt-6-astraQuality 1029Time 1538sCost $5.6995
  3. 1opencode×claude-fable-5-1Quality 1029Time 1918.4sCost $19.8827
  4. 1opencode×claude-opus-5-5Quality 1029Time 1968.1sCost $5.9846
  5. 1opencode×glm-5.3-primeQuality 1029Time 1967sCost $4.7754
  6. 1opencode×gpt-6-astraQuality 1029Time 2024.2sCost $6.3398
  7. 1opencode×qwen3.8-flashQuality 1029Time 1231.7sCost $0.0705
  8. 1opencode×qwen3.8-maxQuality 1029Time 1477.4sCost $0.5318
  9. 9claude_code×glm-5.3Quality 1019Time 1460.9sCost $2.6473
  10. 10claude_code×qwen3.8-maxQuality 1017Time 2570sCost $2.3434
  11. 11codex×glm-5.3-primeQuality 1013Time 795.1sCost $1.3245
  12. 12codex×gpt-6.1-solQuality 1011Time 629.7sCost $0.4992
  13. 13opencode×gpt-6.1-solQuality 1010Time 540.9sCost $0.2253
  14. 14codex×qwen3.8-maxQuality 1010Time 1778.6sCost $1.1965
  15. 15codex×qwen3.8-flashQuality 1009Time 3015.9sCost $0.4022
  16. 16codex×gpt-6-solQuality 1008Time 508.6sCost $0.2762
  17. 17kkimi_cli×kimi-k3Quality 1007Time 1624sCost $3.6929
  18. 18claude_code×glm-5.3-primeQuality 1003Time 1682sCost $4.9994
  19. 19opencode×gpt-6-solQuality 1001Time 1385sCost $0.3287
  20. 20codex×gpt-5.6-solQuality 1000Time 562.2sCost $0.4220