Model comparison
Hy4 preview vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 45.3 on the Noometry Index. Hy4 preview costs 5.3× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
TencentSummary
- They share 3 benchmarks with published results for both. Hy4 preview scores higher in 0 categories and Kimi K3 in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 31.9.
- The biggest single-benchmark swing is NYT Connections (extended): 68.2% for Hy4 preview and 93.6% for Kimi K3.
- Hy4 preview is cheaper at $0.75 / $2.25 per million input/output tokens, against $3 / $15 for Kimi K3.
Side by side
| Hy4 preview | Kimi K3 | |
|---|---|---|
| Provider | Tencent | Moonshot AI |
| Noometry Index | 45.3 | 59.5 |
| Released | 2026-08-28 | 2026-07-16 |
| Weights | Open | Open |
| Context window | 1.05M | 1.05M |
| Max output | 64K | 1.05M |
| Input $ / M tokens | $0.75 | $3 |
| Output $ / M tokens | $2.25 | $15 |
| Results tracked | 3 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Hy4 preview: 51.6 (#38), Kimi K3: 61.0 (#10)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| LMArena WebDev | 1632 | 1654 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| WeirdML | — | 82.6% |
| LMArena Coding | — | 1508 |
| ALE-Bench | — | 1,524 |
Agentic & Tool Use Not comparable
Hy4 preview: —, Kimi K3: 41.8 (#20)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Hy4 preview: 31.9 (#79), Kimi K3: 63.0 (#17)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| NYT Connections (extended) | 68.2% | 93.6% |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Chess Puzzles | — | 39% |
| LMArena Hard Prompts | — | 1496 |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| Epoch Capabilities Index | — | 157.45 |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
Hy4 preview: 55.7 (#42), Kimi K3: 74.2 (#16)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| ProofBench | 75% | 87% |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| OTIS Mock AIME 2024-2025 | — | 97.2% |
| LMArena Math | — | 1491 |
Knowledge Not comparable
Hy4 preview: —, Kimi K3: 63.2 (#21)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| GPQA Diamond | — | 93.1% |
| SimpleQA Verified | — | 50.6% |
| LMArena Expert | — | 1521 |
Multimodal Not comparable
Hy4 preview: —, Kimi K3: 37.8 (#70)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Not comparable
Hy4 preview: —, Kimi K3: 56.3 (#21)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| LMArena Non-English | — | 1466 |
| LMArena Chinese | — | 1529 |
| LMArena French | — | 1491 |
| LMArena German | — | 1488 |
| LMArena Japanese | — | 1487 |
| LMArena Korean | — | 1458 |
| LMArena Russian | — | 1482 |
| LMArena Spanish | — | 1472 |
Instruction Following Not comparable
Hy4 preview: —, Kimi K3: 77.7 (#14)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | — | 1483 |
Long Context Not comparable
Hy4 preview: —, Kimi K3: 45.8 (#29)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| LMArena Longer Query | — | 1494 |
Writing & Preference Not comparable
Hy4 preview: —, Kimi K3: 76.6 (#4)
| Benchmark | Hy4 preview | Kimi K3 |
|---|---|---|
| LMArena Text | — | 1476 |
| LMArena Creative Writing | — | 1454 |
| EQ-Bench Creative Writing | — | 2082 |
| EQ-Bench 4 | — | 1339 |
| LMArena Multi-Turn | — | 1488 |
Frequently asked questions
Is Hy4 preview better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 45.3 on the Noometry Index. Hy4 preview costs 5.3× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Hy4 preview or Kimi K3?
Hy4 preview is cheaper. It lists at $0.75 per million input tokens and $2.25 per million output tokens; Kimi K3 lists at $3 and $15.
Is Hy4 preview or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 51.6 in the Noometry coding category.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do Hy4 preview and Kimi K3 share?
3 benchmarks have published results for both models. Hy4 preview has 3 scored results on Noometry and Kimi K3 has 53.