Hy4 previewvsKimi K3
Across 16 shared benchmarks, Kimi K3 leads overall: Hy4 preview wins 3, Kimi K3 wins 13, with 0 ties and an average score difference of -5.91.
Hy4 preview
腾讯AI实验室 · 2026-08-28 · Reasoning model
Kimi K3
Moonshot AI · 2026-07-16 · Reasoning model
Hy4 preview3 wins(19%)(81%)13 winsKimi K3
Benchmark scores
Grouped by capability, sorted by largest gap within each. 16 shared benchmarks.
AI Agent - Tool Usage
Kimi K3 3/4| Benchmark | Hy4 preview | Kimi K3 | Diff |
|---|---|---|---|
| Terminal-Bench 2.1 | 85.4010 / 47Thinking High (With Tools) | 88.302 / 47Max (With Tools) | -2.90 |
| Toolathlon-Verified | 74.103 / 8Thinking High (With Tools) | 76.502 / 8Max (With Tools) | -2.40 |
| AutomationBench | 32.103 / 10Thinking High (With Tools) | 30.805 / 10Max (With Tools) | +1.30 |
| MCP-Atlas | 83.705 / 40Thinking High (With Tools) | 84.204 / 40Max (With Tools) | -0.50 |
Coding and Software Engineer
Kimi K3 4/4| Benchmark | Hy4 preview | Kimi K3 | Diff |
|---|---|---|---|
| Program Bench | 17.506 / 6Thinking High (With Tools) | 77.801 / 6Max (With Tools) | -60.30 |
| SWE-Marathon | 31.903 / 5Thinking High (With Tools) | 422 / 5Max (With Tools) | -10.10 |
| DeepSWE | 64.3011 / 31Thinking High (With Tools) | 67.505 / 31Max (With Tools) | -3.20 |
| PostTrain Bench | 35.604 / 5Thinking High (With Tools) | 36.603 / 5Max (With Tools) | -1 |
Agent Level Benchmark
Kimi K3 2/3| Benchmark | Hy4 preview | Kimi K3 | Diff |
|---|---|---|---|
| Job Bench | 61.701 / 3Thinking High (With Tools) | 54.303 / 3Max (With Tools) | +7.40 |
| Agents' Last Exam | 22.8013 / 14Thinking High (With Tools) | 28.305 / 14Max (With Tools) | -5.50 |
| APEX-Agents | 37.106 / 6Thinking High (With Tools) | 415 / 6Max (With Tools) | -3.90 |
General Knowledge
Kimi K3 2/2| Benchmark | Hy4 preview | Kimi K3 | Diff |
|---|---|---|---|
| CritPt | 16.902 / 3Thinking High (No Tools) | 23.401 / 3最高（无工具） | -6.50 |
| HLE | 55.4015 / 185Thinking High (With Tools) | 5614 / 185Max (With Tools) | -0.60 |
Productivity Knowledge
Even 2/2| Benchmark | Hy4 preview | Kimi K3 | Diff |
|---|---|---|---|
| GDPval-AA v2 | 1,6788 / 16Thinking High (With Tools) | 1,6867 / 16Max (With Tools) | -8 |
| Office QA Pro | 66.201 / 3Thinking High (With Tools) | 63.302 / 3Max (With Tools) | +2.90 |
General Evaluation
Kimi K3 1/1| Benchmark | Hy4 preview | Kimi K3 | Diff |
|---|---|---|---|
| GPQA Diamond | 92.3024 / 226Thinking High (No Tools) | 93.5013 / 226最高（无工具） | -1.20 |
Specs
| Field | Hy4 preview | Kimi K3 |
|---|---|---|
| Publisher | 腾讯AI实验室 | Moonshot AI |
| Release date | 2026-08-28 | 2026-07-16 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | MoE |
| Parameters | 770B | 2.8T |
| Context length | 1M | 1M |
| Max output | Not available | 1M |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Hy4 preview | Kimi K3 |
|---|---|---|
| Text input | ¥6 / 1M tokens | ¥20 / 1M tokens |
| Text output | ¥18 / 1M tokens | ¥100 / 1M tokens |
| Cache read | ¥0.3 / 1M tokens | ¥2 / 1M tokens |
Summary
- Kimi K3leads in:AI Agent - Tool Usage (3/4), Coding and Software Engineer (4/4), Agent Level Benchmark (2/3), General Knowledge (2/2), General Evaluation (1/1)
- Tied in:Productivity Knowledge
On average across the 16 shared benchmarks, Kimi K3 scores 5.91 higher.
Largest single-benchmark gap: Program Bench — Hy4 preview 17.50 vs Kimi K3 77.80 (-60.30).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.