Claude Haiku 5.5 vs MiMo V2.6 Pro: Benchmark Comparison
Claude Haiku 5.5 has the higher score on 16 of 23 shared benchmarks; MiMo V2.6 Pro leads on 7.
The largest observed score gap is 19.82 pts on MysteryMechanism , where Claude Haiku 5.5 leads.
Reported ±1 standard-error ranges overlap on 14 of 23 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 | Claude Haiku 5.5 | MiMo V2.6 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 57.45% ±1.80 | 57.85% ±1.57 | 0.40 pts | Reported ±1 SE ranges overlap |
| Vals Index | 58.71% ±1.28 | 57.81% ±1.24 | 0.89 pts | Reported ±1 SE ranges overlap |
| Vals Legal Index | 70.27% ±1.83 | 72.14% ±1.83 | 1.87 pts | Reported ±1 SE ranges overlap |
| HLAB-Vals-1.1 | 4.17% ±1.52 | 13.75% ±2.84 | 9.58 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 43.27% ±3.44 | 47.12% ±3.47 | 3.85 pts | Reported ±1 SE ranges overlap |
| EMB v1.1 | 65.48% ±3.69 | 67.35% ±2.90 | 1.87 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 65.37% ±2.06 | 64.26% ±2.02 | 1.11 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 31.92% ±3.35 | 32.39% ±3.36 | 0.47 pts | Reported ±1 SE ranges overlap |
| MedCode | 51.54% ±2.17 | 44.97% ±2.10 | 6.57 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 89.54% ±1.92 | 88.31% ±1.94 | 1.23 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 89.00% ±3.15 | 70.00% ±4.61 | 19.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 35.13% ±3.21 | 15.31% ±2.42 | 19.82 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 10.00% ±3.61 | 2.86% ±2.01 | 7.14 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 45.78% ±3.42 | 45.05% ±3.40 | 0.73 pts | Reported ±1 SE ranges overlap |
| Code Migration v1.1 | 52.36% ±3.49 | 48.79% ±3.88 | 3.57 pts | Reported ±1 SE ranges overlap |
| IOI | 47.28% ±10.72 | 39.33% ±2.41 | 7.95 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 2.50% ±1.11 | 0.50% ±0.50 | 2.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 43.96% ±2.27 | 41.25% ±2.25 | 2.71 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 36.36% ±2.31 | 31.31% ±3.07 | 5.05 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 90.44% ±1.73 | 85.22% ±3.39 | 5.22 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 75.48% ±5.21 | 72.86% ±5.50 | 2.62 pts | Reported ±1 SE ranges overlap |
| SRE Bench | 5.72% ±1.44 | 3.05% ±1.06 | 2.67 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 63.19% ±1.25 | 68.94% ±1.20 | 5.75 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Haiku 5.5 average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 23.72% | 30.43% |
| Finance | 54.26% | 54.66% |
| Healthcare | 70.54% | 66.64% |
| Math | 89.00% | 70.00% |
| Science | 22.57% | 9.09% |
| Education | 45.78% | 45.05% |
| Coding | 45.48% | 41.07% |
| Cyber | 40.60% | 37.95% |
| Social Mobility | 63.19% | 68.94% |
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 | Claude Haiku 5.5 cost | MiMo V2.6 Pro cost | Claude Haiku 5.5 latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Finance Index | $1.39 | $0.22 | 33m31s | 30m48s |
| Vals Index | $3.17 | $0.43 | 49m48s | 1h13m |
| Vals Legal Index | $2.35 | $0.18 | 58m22s | 29m02s |
| HLAB-Vals-1.1 | $4.26 | $0.22 | 43m41s | 31m02s |
| Legal Research Bench | $1.63 | $0.18 | 1h22m | 30m16s |
| EMB v1.1 | $1.36 | $0.33 | 19m56s | 55m14s |
| Finance Agent v2.1 | $1.64 | $0.18 | 31m05s | 11m22s |
| Tax Agent Bench | $1.16 | $0.13 | 49m34s | 25m48s |
| MedCode | N/A | N/A | 3m46s | 2m57s |
| MedScribe | N/A | N/A | 5m23s | 2m53s |
| ProofBench v1.1 | $0.17 | $0.25 | 6m58s | 1h02m |
| MysteryMechanism | $0.54 | $0.15 | 13m25s | 44m00s |
| Terminal-Bench-Science | $4.52 | $0.62 | 1h34m | 3h12m |
| SAGE | N/A | N/A | 2m23s | 3m40s |
| Code Migration v1.1 | $6.20 | $0.99 | 1h26m | 3h28m |
| IOI | $0.83 | $0.84 | 12m34s | 2h26m |
| ProgramBench | $6.83 | $0.73 | 1h13m | 3h00m |
| SWE-bench Multimodal | $0.70 | $0.12 | 16m16s | 46m03s |
| Terminal-Bench 4.0 | $3.12 | $0.50 | 51m56s | 2h42m |
| Vibe Code Bench v1.1 | $6.07 | $1.04 | 34m36s | 1h01m |
| CyberBench v1.1 | $1.25 | $0.09 | 32m39s | 27m41s |
| SRE Bench | $3.14 | $0.53 | 57m56s | 2h23m |
| Public Benefits Bench v1.1 | $0.53 | $0.08 | 36m32s | 37m14s |
Results available only for Claude Haiku 5.5
- Vals Science Index
- ClaimBench
- BioMysteryBench
- Vibe Code Bench 1-100
Results available only for MiMo V2.6 Pro
None.