Leaderboard

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

Data analysis

Spreadsheet analysis

Pairings
49
Tasks
15
Order
Highest quality first

Quality and cost

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

↑ Quality
6008001000120014001600

On the line

  1. deepseek_harness × deepseek-flashQuality922Cost$0.0041
  2. opencode × deepseek-flashQuality1039Cost$0.0105
  3. codex × deepseek-flashQuality1092Cost$0.0130
  4. codex × gpt-6-solQuality1193Cost$0.0610
  5. opencode × kimi-k3Quality1212Cost$0.0707
  6. codex × gpt-6.1-solQuality1286Cost$0.0735
  7. opencode × gpt-6.1-solQuality1287Cost$0.0809
  8. claude_code × claude-fable-5-1Quality1400Cost$0.1112

All pairings

  1. 1claude_code×claude-fable-5-1Quality 1400Time 70.7sCost $0.1112
  2. 2opencode×claude-fable-5-1Quality 1382Time 103.5sCost $0.1499
  3. 3claude_code×claude-opus-5-5Quality 1351Time 68.3sCost $0.1563
  4. 4opencode×claude-opus-5-5Quality 1350Time 88.8sCost $0.1394
  5. 5opencode×gpt-5.6-solQuality 1330Time 167.7sCost $0.1662
  6. 6codex×gpt-6-astraQuality 1329Time 168.9sCost $0.1656
  7. 7opencode×gpt-6-astraQuality 1291Time 393.3sCost $0.2760
  8. 8opencode×gpt-6.1-solQuality 1287Time 150.6sCost $0.0809
  9. 9codex×gpt-6.1-solQuality 1286Time 107.8sCost $0.0735
  10. 10opencode×gpt-6-solQuality 1278Time 130.8sCost $0.0748
  11. 11codex×gpt-5.6-solQuality 1265Time 175.4sCost $0.1145
  12. 12opencode×gpt-5.6-terraQuality 1218Time 123.7sCost $0.0868
  13. 13opencode×kimi-k3Quality 1212Time 122.1sCost $0.0707
  14. 14codex×gpt-6-solQuality 1193Time 94.6sCost $0.0610
  15. 15codex×glm-5.3-primeQuality 1123Time 115.9sCost $0.1203
  16. 16codex×deepseek-flashQuality 1092Time 92.7sCost $0.0130
  17. 17opencode×claude-sonnet-5Quality 1081Time 113.8sCost $0.0564
  18. 18codex×kimi-k3Quality 1062Time 167.5sCost $0.0725
  19. 19opencode×deepseek-flashQuality 1039Time 96.4sCost $0.0105
  20. 20codex×gpt-5.6-terraQuality 1037Time 106.7sCost $0.0654