List your company
ResearchFormerly WizWand

Task Fulfillment on MCPBench

4.38Task Fulfillment ScoreKimi-K2 Thinking
2.2482.80153.3553.9085Mar 5, 2026
Updated 8mo ago

Evaluation results

MethodLinks
Kimi-K2 ThinkingExecution Strategy=All Tools Loaded2026.034.382023,768
Qwen3-4B-InstructExecution Strategy=Iterative Tool Loading, Programmatic Tool Calling (PTC)=true, Reinforcement Learning=RL w/ Qwen3-30B J. (Rubrics)2026.034.151813,400
Kimi-K2 ThinkingExecution Strategy=Iterative Server Loading2026.034.112721,747
Qwen3-4B-InstructExecution Strategy=Iterative Tool Loading, Reinforcement Learning=RL w/ Qwen3-30B J. (Rubrics)2026.034.032011,151
Qwen3-4B-InstructExecution Strategy=Iterative Tool Loading, Programmatic Tool Calling (PTC)=true, Reinforcement Learning=RL w/ Qwen3-30B J. (Generic)2026.033.912012,951
Qwen3-4B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ Qwen3-30B Judge (Rubrics)2026.033.871911,142
Kimi-K2 ThinkingExecution Strategy=Iterative Tool Loading2026.033.622818,290
Qwen3-4B-InstructExecution Strategy=Iterative Tool Loading, Reinforcement Learning=RL w/ Qwen3-30B J. (Generic)2026.033.482312,815
Qwen3-4B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ GPT-4o Judge (Rubrics)2026.033.431810,654
Qwen3-4B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ GPT-4o Judge (Generic)2026.033.25179,372
Qwen2.5-7B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ Qwen3-30B Judge (Rubrics)2026.033.182113,075
Qwen2.5-7B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ GPT-4o Judge (Rubrics)2026.033.131911,882
Qwen3-4B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ Qwen3-30B Judge (Generic)2026.033.122111,530
Qwen2.5-7B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ GPT-4o Judge (Generic)2026.033.021910,307
Qwen3-4B-InstructExecution Strategy=Iterative Tool Loading, Programmatic Tool Calling (PTC)=true2026.032.941813,462
Qwen2.5-7B-InstructExecution Strategy=Iterative Server Loading, Reinforcement Learning=RL w/ Qwen3-30B Judge (Generic)2026.032.842412,201
Qwen3-4B-InstructExecution Strategy=Iterative Server Loading2026.032.73249,152
Qwen3-4B-InstructExecution Strategy=Iterative Tool Loading2026.032.36209,045
Qwen2.5-7B-InstructExecution Strategy=Iterative Server Loading2026.032.332511,192