Muse Spark 1.3 vs MiMo V2.6 Flash: Benchmark Comparison
Muse Spark 1.3 has the higher score on 5 of 14 shared benchmarks; MiMo V2.6 Flash leads on 7.
The largest observed score gap is 13.64 pts on Terminal-Bench 4.0 , where MiMo V2.6 Flash leads.
Reported ±1 standard-error ranges overlap on 9 of 14 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 | Muse Spark 1.3 | MiMo V2.6 Flash | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 53.20% ±1.07 | 53.23% ±1.10 | 0.03 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 22.92% ±3.42 | 11.25% ±2.54 | 11.67 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 40.87% ±3.42 | 37.98% ±3.37 | 2.88 pts | Reported ±1 SE ranges overlap |
| EMB | 62.71% ±2.96 | 65.46% ±2.71 | 2.75 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 58.90% ±0.44 | 56.28% ±0.46 | 2.62 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 71.93% ±2.92 | 59.90% ±3.13 | 12.03 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 55.00% ±5.00 | 63.00% ±4.85 | 8.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench Science | 4.29% ±2.44 | 4.29% ±2.44 | 0.00 pts | Reported ±1 SE ranges overlap |
| Code Migration | 27.58% ±4.07 | 40.93% ±4.35 | 13.35 pts | Reported ±1 SE ranges do not overlap |
| IOI | 43.94% ±1.45 | 47.67% ±4.79 | 3.72 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 0.50% ±0.50 | 0.50% ±0.50 | 0.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 10.61% ±1.75 | 24.24% ±1.51 | 13.64 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 82.86% ±2.90 | 78.96% ±4.04 | 3.90 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 69.41% ±5.76 | 75.36% ±5.42 | 5.95 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Muse Spark 1.3 average | MiMo V2.6 Flash average |
|---|---|---|
| Legal | 31.89% | 24.62% |
| Finance | 64.52% | 60.55% |
| Math | 55.00% | 63.00% |
| Science | 4.29% | 4.29% |
| Coding | 33.10% | 38.46% |
| Cyber | 69.41% | 75.36% |
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 | Muse Spark 1.3 cost | MiMo V2.6 Flash cost | Muse Spark 1.3 latency | MiMo V2.6 Flash latency |
|---|---|---|---|---|
| Vals Index | $3.38 | $0.20 | 32m42s | 58m02s |
| Harvey's Legal Agent Benchmark | $2.76 | $0.09 | 16m39s | 17m12s |
| Legal Research Bench | $0.65 | $0.07 | 8m51s | 15m20s |
| EMB | $3.55 | $0.11 | 22m55s | 25m22s |
| Finance Agent (v2) | $0.74 | $0.07 | 5m42s | 7m02s |
| Tax Agent Bench | $0.32 | $0.05 | 8m25s | 17m19s |
| ProofBench v1.1 | $0.43 | $0.12 | 8m54s | 1h03m |
| Terminal-Bench Science | $4.41 | $0.24 | 1h08m | 3h59m |
| Code Migration | $4.62 | $0.49 | 1h08m | 3h12m |
| IOI | $2.36 | $0.22 | 29m51s | 1h49m |
| ProgramBench | $11.56 | $0.47 | 1h09m | 3h22m |
| Terminal-Bench 4.0 | $7.04 | $0.22 | 1h59m | 2h25m |
| Vibe Code Bench v1.1 | $2.10 | $0.56 | 12m54s | 50m13s |
| CyberBench v1.1 | $4.26 | $0.05 | 24m53s | 28m22s |
Results available only for Muse Spark 1.3
None.
Results available only for MiMo V2.6 Flash
- MedCode
- MedScribe
- BioMysteryBench
- MysteryMechanism
- SAGE
- SRE Bench
- Public Benefits Bench v1.1