Harness
claude_code
claude_code is strongest with claude-opus-5-5, 1st overall.
Best overall rank1st
Best partner
claude-opus-5-5
claude-opus-5-5Strongest inPresentations
First places16 tasks
Pairings evaluated10 of 19 models
Time and cost, best pairingTime752.6sCost$3.4097
Who claude_code works best with
Valid evaluations covered 10 of 19 models. The other 9 have not been evaluated yet.
- claude-opus-5-51172#1
glm-5.3-prime1131#3- glm-5.31060#9
qwen3.8-flash1058#10- claude-fable-5-11038#14
- qwen3.8-max1021#17
- claude-sonnet-5973#29
- qwen3.7-plus955#32
MiniMax-M2.5854#38- claude-haiku-4-5-20251001852#39
kimi-k3—Not evaluated
gpt-6.1-sol—Not evaluated- gpt-6-astra—Not evaluated
- gpt-5.6-terra—Not evaluated
- gpt-5.6-sol—Not evaluated
- gpt-6-sol—Not evaluated
deepseek-flash—Not evaluated- kimi-k2.7-code—Not evaluated
- deepseek-v4-pro—Not evaluated
Places across 24 tasks
- 161st
- 42nd
- 13rd
- 14th
- 05th
- 26th and below
Most often 1st, in 16 tasks.
Rank in each kind of work
Each row is a kind of work and each dot is one pairing's place in it; further left is better. Hover a dot for its name and place, or pick a name below to follow one pairing.
Where it places in each task
The number in each cell is a place: where that pairing ranked in that task, 1 being best, so 3 means 3rd. Hover a number to see how many pairings were ranked there. A dash means the pairing was not run on that task.
1st2nd3rd4th5th6th and below