Qwen 3.8 Max vs Kimi K3: Benchmark Comparison
Qwen 3.8 Max has the higher score on 8 of 25 shared benchmarks; Kimi K3 leads on 17.
The largest observed score gap is 46.67 pts on CyberBench v1.1 , where Kimi K3 leads.
Reported ±1 standard-error ranges overlap on 12 of 24 shared benchmarks with comparable uncertainty data. This is not a pairwise statistical significance test.
Shared benchmark results
Scores and reported standard errors come from the latest visible Vals benchmark version available for both models.
| Benchmark | Qwen 3.8 Max | Kimi K3 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 21.76% | 20.88% | 0.88 pts | Uncertainty comparison unavailable |
| Legal Research Bench | 47.60% ±3.47 | 46.15% ±3.46 | 1.44 pts | Reported ±1 SE ranges overlap |
| LegalBench | 83.61% ±0.42 | 86.21% ±0.41 | 2.60 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 65.99% ±2.69 | 71.26% ±2.28 | 5.28 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 63.99% ±0.94 | 66.34% ±0.91 | 2.35 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 32.09% ±3.37 | 33.37% ±3.39 | 1.27 pts | Reported ±1 SE ranges overlap |
| TaxEval v2 | 75.55% ±0.84 | 75.72% ±0.84 | 0.16 pts | Reported ±1 SE ranges overlap |
| MedCode | 40.67% ±2.03 | 49.36% ±2.20 | 8.69 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 84.95% ±2.00 | 88.05% ±1.98 | 3.10 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 58.00% ±4.96 | 87.00% ±3.38 | 29.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 12.86% ±4.03 | 1.43% ±1.43 | 11.43 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 93.69% ±1.22 | 92.93% ±1.31 | 0.76 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 88.60% ±0.31 | 87.97% ±0.32 | 0.63 pts | Reported ±1 SE ranges overlap |
| MMMU Pro | 88.03% ±0.78 | 88.15% ±0.78 | 0.12 pts | Reported ±1 SE ranges overlap |
| SAGE | 51.25% ±3.42 | 52.78% ±3.41 | 1.53 pts | Reported ±1 SE ranges overlap |
| IOI | 68.89% ±5.20 | 48.94% ±9.82 | 19.94 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 87.85% ±0.95 | 87.19% ±0.97 | 0.67 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 0.00% ±0.00 | 2.00% ±0.99 | 2.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 85.60% ±1.57 | 93.40% ±1.11 | 7.80 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 42.50% ±2.26 | 44.17% ±2.27 | 1.67 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 34.34% ±3.94 | 17.17% ±0.51 | 17.17 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench 1-100 | 12.83% ±3.18 | 18.24% ±3.92 | 5.41 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 64.70% ±5.36 | 84.97% ±2.75 | 20.27 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 28.57% ±3.31 | 75.24% ±5.56 | 46.67 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 67.12% ±1.22 | 68.20% ±1.21 | 1.08 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Qwen 3.8 Max average | Kimi K3 average |
|---|---|---|
| Legal | 65.60% | 66.18% |
| Finance | 59.41% | 61.67% |
| Healthcare | 62.81% | 68.70% |
| Math | 58.00% | 87.00% |
| Science | 12.86% | 1.43% |
| Academic | 90.11% | 89.68% |
| Education | 51.25% | 52.78% |
| Coding | 49.59% | 49.51% |
| Cyber | 28.57% | 75.24% |
| Social Mobility | 67.12% | 68.20% |
Cost and latency
Cost per test appears only where the benchmark reports it for both models. Latency is the measured completion time for that benchmark.
| Benchmark | Qwen 3.8 Max cost | Kimi K3 cost | Qwen 3.8 Max latency | Kimi K3 latency |
|---|---|---|---|---|
| Vals RSI Index | $113.78 | $136.82 | 90h00m | 90h00m |
| Legal Research Bench | $2.49 | $3.47 | 1h01m | 12m40s |
| LegalBench | N/A | N/A | 36.37s | 7.11s |
| EMB v1.1 | $2.66 | $3.07 | 57m47s | 12m21s |
| MortgageTax | N/A | N/A | 47.50s | 78.94s |
| Tax Agent Bench | $1.55 | $2.79 | 40m36s | 40m15s |
| TaxEval v2 | N/A | N/A | 4m26s | 2m25s |
| MedCode | N/A | N/A | 6m18s | 36.99s |
| MedScribe | N/A | N/A | 4m11s | 38.24s |
| ProofBench v1.1 | $1.50 | $1.65 | 31m08s | 34m31s |
| Terminal-Bench-Science | $16.68 | $21.30 | 3h24m | 4h49m |
| GPQA Diamond | N/A | N/A | 4m27s | 2m02s |
| MMLU Pro | N/A | N/A | 63.58s | 38.72s |
| MMMU Pro | N/A | N/A | 43.85s | 106.52s |
| SAGE | N/A | N/A | 2m13s | 47.30s |
| IOI | $9.47 | $14.17 | 1h47m | 4h18m |
| LiveCodeBench | N/A | N/A | 5m38s | 3m20s |
| ProgramBench | $12.36 | $70.48 | 5h54m | 5h42m |
| SWE-bench | $1.13 | $0.76 | 41m42s | 10m19s |
| SWE-bench Multimodal | $0.99 | $1.65 | 18m44s | 22m45s |
| Terminal-Bench 4.0 | $10.63 | $12.48 | 2h28m | 3h09m |
| Vibe Code Bench 1-100 | $9.68 | $9.40 | 1h36m | 1h48m |
| Vibe Code Bench v1.1 | $8.24 | $10.01 | 2h49m | 16m39s |
| CyberBench v1.1 | $0.78 | $2.13 | 9m40s | 30m44s |
| Public Benefits Bench v1.1 | $0.99 | $0.91 | 40m44s | 7m19s |
Results available only for Qwen 3.8 Max
- MysteryMechanism
- SkillsBench
Results available only for Kimi K3
- Vals Finance Index
- Vals Index
- Vals Inference Provider Index
- Vals Legal Index
- HLAB-Vals-1.1
- Finance Agent v2.1
- BioMysteryBench
- Code Migration v1.1
- Time Horizon Index: KSP