MiMo V2.6 Flash vs GLM 5.3: Benchmark Comparison
MiMo V2.6 Flash has the higher score on 3 of 21 shared benchmarks; GLM 5.3 leads on 18.
The largest observed score gap is 20.78 pts on IOI , where GLM 5.3 leads.
Reported ±1 standard-error ranges overlap on 11 of 21 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 | MiMo V2.6 Flash | GLM 5.3 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 57.15% ±1.39 | 60.54% ±1.52 | 3.39 pts | Reported ±1 SE ranges do not overlap |
| Vals Index | 54.83% ±1.11 | 59.73% ±1.11 | 4.90 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 68.99% ±1.77 | 74.58% ±1.84 | 5.59 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 13.33% ±2.82 | 14.58% ±2.86 | 1.25 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 37.98% ±3.37 | 49.04% ±3.48 | 11.06 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 68.88% ±2.27 | 69.11% ±2.65 | 0.23 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 63.32% ±2.08 | 65.00% ±2.01 | 1.69 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 28.51% ±3.14 | 39.70% ±3.52 | 11.19 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 41.06% ±2.03 | 42.86% ±2.11 | 1.81 pts | Reported ±1 SE ranges overlap |
| MedScribe | 85.28% ±1.99 | 88.81% ±2.00 | 3.53 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 63.00% ±4.85 | 49.00% ±5.02 | 14.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 21.62% ±2.77 | 22.97% ±2.83 | 1.35 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench-Science | 4.29% ±2.44 | 5.71% ±2.79 | 1.43 pts | Reported ±1 SE ranges overlap |
| Code Migration v1.1 | 39.55% ±3.54 | 55.07% ±2.12 | 15.52 pts | Reported ±1 SE ranges do not overlap |
| IOI | 47.67% ±4.79 | 68.44% ±7.48 | 20.78 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.50% ±0.50 | 1.50% ±0.86 | 1.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 24.24% ±1.51 | 38.89% ±1.82 | 14.65 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 78.96% ±4.04 | 78.13% ±3.94 | 0.83 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 75.36% ±5.42 | 72.08% ±5.41 | 3.27 pts | Reported ±1 SE ranges overlap |
| SRE Bench | 4.20% ±1.24 | 9.16% ±1.79 | 4.96 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 67.59% ±1.22 | 68.54% ±1.21 | 0.95 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | MiMo V2.6 Flash average | GLM 5.3 average |
|---|---|---|
| Legal | 25.66% | 31.81% |
| Finance | 53.57% | 57.94% |
| Healthcare | 63.17% | 65.84% |
| Math | 63.00% | 49.00% |
| Science | 12.95% | 14.34% |
| Coding | 38.18% | 48.41% |
| Cyber | 39.78% | 40.62% |
| Social Mobility | 67.59% | 68.54% |
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 | MiMo V2.6 Flash cost | GLM 5.3 cost | MiMo V2.6 Flash latency | GLM 5.3 latency |
|---|---|---|---|---|
| Vals Finance Index | $0.07 | $2.21 | 19m26s | 35m44s |
| Vals Index | $0.22 | $8.52 | 1h07m | 1h04m |
| Vals Legal Index | $0.07 | $2.68 | 18m20s | 44m51s |
| HLAB-Vals-1.1 | $0.09 | $3.99 | 22m23s | 31m51s |
| Legal Research Bench | $0.07 | $2.24 | 15m20s | 51m50s |
| EMB v1.1 | $0.11 | $3.79 | 25m22s | 36m12s |
| Finance Agent v2.1 | $0.07 | $1.03 | 15m36s | 20m06s |
| Tax Agent Bench | $0.05 | $1.80 | 17m19s | 50m53s |
| MedCode | N/A | N/A | 93.37s | 2m49s |
| MedScribe | N/A | N/A | 74.48s | 2m02s |
| ProofBench v1.1 | $0.12 | $2.08 | 1h03m | 42m23s |
| MysteryMechanism | $0.06 | $1.53 | 35m08s | 30m42s |
| Terminal-Bench-Science | $0.24 | $15.63 | 3h59m | 2h45m |
| Code Migration v1.1 | $0.53 | $34.91 | 4h10m | 2h58m |
| IOI | $0.22 | $7.67 | 1h49m | 1h46m |
| ProgramBench | $0.47 | $21.96 | 3h22m | 4h16m |
| Terminal-Bench 4.0 | $0.22 | $9.37 | 2h25m | 1h23m |
| Vibe Code Bench v1.1 | $0.56 | $12.45 | 50m13s | 1h04m |
| CyberBench v1.1 | $0.05 | $2.69 | 28m22s | 30m18s |
| SRE Bench | $0.40 | $27.47 | 3h22m | 2h48m |
| Public Benefits Bench v1.1 | $0.03 | $0.95 | 20m56s | 35m51s |
Results available only for MiMo V2.6 Flash
- Vals Science Index
- BioMysteryBench
- SAGE
Results available only for GLM 5.3
- Vals RSI Index
- LegalBench
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- LiveCodeBench
- SkillsBench
- SWE-bench
- Vibe Code Bench 1-100