Compare
Kimi K2.5 vs Tencent Hy3 preview
Add, remove, or swap models to compare them side by side.
Scores
Intelligence (ECI)148—
Coding44—
Math53—
Reasoning & Knowledge36—
Agentic & Tools39—
Specifications
DeveloperMoonshotTencent
FamilyKimi—
ReleasedFeb 2, 2026Apr 23, 2026
Parameters1T295B
AvailabilityOpen weights (unrestricted)Open weights (restricted use)
Context window262K—
Price — $/M input$0.38—
Price — $/M output$2.02—
Inputstext, image—
Outputstext—
Benchmarks
AIME 2024/202592%—
APEX14%—
ARC-AGI65%—
ARC-AGI-212%—
FrontierMath28%—
FrontierMath Tier 44%—
GPQA Diamond88%—
Humanity's Last Exam24%—
SimpleBench47%—
SimpleQA Verified34%—
SWE-bench Verified74%—
Terminal-Bench43%—
WebDev Arena1431—
WeirdML46%—
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Popular comparisons
- Claude Opus 4.8 vs GPT-5.5
- Gemini 3.1 Pro vs GPT-5.5
- GLM-5.1 vs GPT-5.5
- GPT-5.5 vs Kimi K2.5
- GPT-5.5 vs Grok 4
- DeepSeek-V3.2 vs GPT-5.5
- Claude Opus 4.8 vs Gemini 3.1 Pro
- Claude Opus 4.8 vs GLM-5.1
- Claude Opus 4.8 vs Kimi K2.5
- Claude Opus 4.8 vs Grok 4
- Claude Opus 4.8 vs DeepSeek-V3.2
- Gemini 3.1 Pro vs GLM-5.1
- Gemini 3.1 Pro vs Kimi K2.5
- Gemini 3.1 Pro vs Grok 4
- DeepSeek-V3.2 vs Gemini 3.1 Pro
- GLM-5.1 vs Kimi K2.5
- GLM-5.1 vs Grok 4
- DeepSeek-V3.2 vs GLM-5.1
- Grok 4 vs Kimi K2.5
- DeepSeek-V3.2 vs Kimi K2.5
- DeepSeek-V3.2 vs Grok 4
- GPT-5 vs GPT-5.5
- GLM-5 vs GLM-5.1
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.