Benchmarks of KAT-Coder-Pro V117
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding3
Agents2
Documents1
Legacy · saturated or no longer reported4
No benchmark matches
—per task?
API price not publicprice
—output speed?
256Kcontext
Rankingin non-thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow+16%—$0.005Function calling only with none · Chat Completions+3
OpenMiMo v2.6 Prothinking×2.5—$0.13Pass back reasoning_content+3BetterGPT-5.6 Lunalow+13%—$0.01BetterMiMo v2.5thinkingOpen+17%—$0.019
BetterClaude Haiku 5.5low+58%—$0.024Remove sampling parameters+5ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow+16%—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+13%—$0.01MiMo v2.5thinkingOpen+17%—$0.019Claude Haiku 5.5low+58%—$0.024Remove sampling parameters+5MiMo v2.5 ProthinkingOpen+40%—$0.054ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen×2.5—$0.13Pass back reasoning_content+3
DeepSeek V4.1 Flash 1017maxOpen×2.4—$0.26Pass back reasoning_content · Chat Completions+4
GLM 5.3maxOpen×2.4—$2.01Reasoning cannot be disabled+3
Step 5 PreviewthinkingOpen×2.4—$1.03web_search not available · Chat Completions+2
Kimi K3maxOpen×2.3—$2.00Sampling parameters are fixed+3History
2 versionsBenchmarks
all categoriesChat & Knowledge
Humanity's Last Exam33.6#69
AA Omniscience Index-37.7#169
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈18.6#128
AA Omniscience · Accuracy17.7#191
AA-LCR (long context) 1.138.7#202
GPQA Diamond76.4#140
AIME 2025 · AA archived94.7#5
Coding
LiveCodeBench · AA archived74.7#30
Agents & Tool Use
GDPval-AA (work tasks) 2.111.6#109
AA-Briefcase 1.1 · Elo601.4#91
τ²-bench · AA88.6#48
Writing & Language
IFBench (instructions)68.4#57
FAQ
How much does KAT-Coder-Pro V1 cost?›
There is no per-unit API price: API price not public.
Is KAT-Coder-Pro V1 free to use?›
There is no free API.
How good is KAT-Coder-Pro V1, and what is it best at?›
#69 of 334 in Chat & Knowledge, #89 of 97 in Coding, #109 of 175 in Agents & Tool Use, #57 of 245 in Writing & Language. Strongest at: Tool calling, Reasoning, Instructions.
Is KAT-Coder-Pro V1 better than MiniMax M2?›
On AA Intelligence Index 4.3.2, KAT-Coder-Pro V1 scores 18.6 vs 18.6 — about the same.
What is the context window of KAT-Coder-Pro V1?›
256K tokens.
When was KAT-Coder-Pro V1 released? Is there a newer version?›
Released 2025-11-11. It has been superseded by KAT Coder Pro V2.
Is KAT-Coder-Pro V1 open source?›
No, KAT-Coder-Pro V1 is a closed model available through an API. The closest open-weights option here is MiMo v2.6 Pro.
Can KAT-Coder-Pro V1 read images?›
No, it accepts text only.
What is KAT-Coder-Pro V1?›
KAT-Coder-Pro V1 is a closed language model from KwaiKAT, released Nov 2025.
What are the best alternatives to KAT-Coder-Pro V1?›
GPT-6 Luna (better), MiMo v2.6 Pro (open weights).