Muse Spark 1.3 vs MiMo V2.6 Pro: Benchmark Comparison
Muse Spark 1.3 has the higher score on 5 of 14 shared benchmarks; MiMo V2.6 Pro leads on 8.
The largest observed score gap is 20.71 pts on Terminal-Bench 4.0 , where MiMo V2.6 Pro leads.
Reported ±1 standard-error ranges overlap on 7 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 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 53.20% ±1.07 | 55.20% ±1.18 | 2.00 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 22.92% ±3.42 | 10.83% ±2.45 | 12.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 40.87% ±3.42 | 47.12% ±3.47 | 6.25 pts | Reported ±1 SE ranges overlap |
| EMB | 62.71% ±2.96 | 62.86% ±3.14 | 0.15 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 58.90% ±0.44 | 57.34% ±0.57 | 1.56 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 71.93% ±2.92 | 64.94% ±3.21 | 7.00 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 55.00% ±5.00 | 70.00% ±4.61 | 15.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 4.29% ±2.44 | 2.86% ±2.01 | 1.43 pts | Reported ±1 SE ranges overlap |
| Code Migration | 27.58% ±4.07 | 43.01% ±4.32 | 15.43 pts | Reported ±1 SE ranges do not overlap |
| IOI | 43.94% ±1.45 | 39.33% ±2.41 | 4.61 pts | Reported ±1 SE ranges do not 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 | 31.31% ±3.07 | 20.71 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 82.86% ±2.90 | 85.22% ±3.39 | 2.36 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 69.41% ±5.76 | 72.86% ±5.50 | 3.45 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Muse Spark 1.3 average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 31.89% | 28.97% |
| Finance | 64.52% | 61.71% |
| Math | 55.00% | 70.00% |
| Science | 4.29% | 2.86% |
| Coding | 33.10% | 39.88% |
| Cyber | 69.41% | 72.86% |
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 Pro cost | Muse Spark 1.3 latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Index | $3.38 | $0.41 | 32m42s | 1h07m |
| Harvey's Legal Agent Benchmark | $2.76 | $0.22 | 16m39s | 25m49s |
| Legal Research Bench | $0.65 | $0.18 | 8m51s | 30m16s |
| EMB | $3.55 | $0.33 | 22m55s | 55m14s |
| Finance Agent (v2) | $0.74 | $0.20 | 5m42s | 10m48s |
| Tax Agent Bench | $0.32 | $0.13 | 8m25s | 25m48s |
| ProofBench v1.1 | $0.43 | $0.25 | 8m54s | 1h02m |
| Terminal-Bench Science | $4.41 | $0.62 | 1h08m | 3h12m |
| Code Migration | $4.62 | $0.69 | 1h08m | 2h49m |
| IOI | $2.36 | $0.84 | 29m51s | 2h26m |
| ProgramBench | $11.56 | $0.73 | 1h09m | 3h00m |
| Terminal-Bench 4.0 | $7.04 | $0.50 | 1h59m | 2h42m |
| Vibe Code Bench v1.1 | $2.10 | $1.04 | 12m54s | 1h01m |
| CyberBench v1.1 | $4.26 | $0.09 | 24m53s | 27m41s |
Results available only for Muse Spark 1.3
None.
Results available only for MiMo V2.6 Pro
- MedCode
- MedScribe
- MysteryMechanism
- SAGE
- SRE Bench
- Public Benefits Bench v1.1