DeepSeek V4.1 Flash vs Grok 4.6: Benchmark Comparison
DeepSeek V4.1 Flash has the higher score on 8 of 23 shared benchmarks; Grok 4.6 leads on 15.
The largest observed score gap is 18.98 pts on SAGE , where DeepSeek V4.1 Flash leads.
Reported ±1 standard-error ranges overlap on 12 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 | DeepSeek V4.1 Flash | Grok 4.6 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 51.32% ±1.13 | 52.09% ±1.15 | 0.77 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 6.67% ±1.83 | 15.83% ±2.65 | 9.17 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 41.35% ±3.42 | 48.08% ±3.47 | 6.73 pts | Reported ±1 SE ranges overlap |
| LegalBench | 83.28% ±0.46 | 86.31% ±0.42 | 3.03 pts | Reported ±1 SE ranges do not overlap |
| EMB | 57.21% ±3.23 | 62.73% ±3.08 | 5.52 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.48% ±0.39 | 53.68% ±0.67 | 0.20 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 62.46% ±3.15 | 70.79% ±3.00 | 8.33 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 41.17% ±2.04 | 44.71% ±2.26 | 3.54 pts | Reported ±1 SE ranges overlap |
| MedScribe | 85.50% ±1.92 | 86.53% ±1.96 | 1.03 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 54.00% ±5.01 | 51.00% ±5.02 | 3.00 pts | Reported ±1 SE ranges overlap |
| BioMysteryBench | 67.78% ±1.11 | 72.22% ±0.00 | 4.44 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 21.17% ±2.75 | 30.63% ±3.10 | 9.46 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 0.00% ±0.00 | 11.43% ±3.83 | 11.43 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 47.88% ±3.44 | 28.90% ±3.08 | 18.98 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 45.62% ±4.29 | 44.57% ±4.48 | 1.05 pts | Reported ±1 SE ranges overlap |
| IOI | 40.28% ±2.56 | 47.61% ±2.09 | 7.33 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.50% ±0.50 | 1.00% ±0.70 | 0.50 pts | Reported ±1 SE ranges overlap |
| SkillsBench | 69.80% ±3.88 | 55.77% ±4.84 | 14.04 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 19.70% ±1.75 | 17.17% ±1.34 | 2.52 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench 1-100 | 16.38% ±3.19 | 14.75% ±3.17 | 1.63 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 84.74% ±2.89 | 76.24% ±3.82 | 8.50 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 73.69% ±5.48 | 67.68% ±5.87 | 6.01 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 64.28% ±1.25 | 66.85% ±1.23 | 2.57 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | DeepSeek V4.1 Flash average | Grok 4.6 average |
|---|---|---|
| Legal | 43.76% | 50.07% |
| Finance | 57.72% | 62.40% |
| Healthcare | 63.34% | 65.62% |
| Math | 54.00% | 51.00% |
| Science | 29.65% | 38.09% |
| Education | 47.88% | 28.90% |
| Coding | 39.57% | 36.73% |
| Cyber | 73.69% | 67.68% |
| Social Mobility | 64.28% | 66.85% |
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 | DeepSeek V4.1 Flash cost | Grok 4.6 cost | DeepSeek V4.1 Flash latency | Grok 4.6 latency |
|---|---|---|---|---|
| Vals Index | $0.33 | $4.49 | 27m53s | 37m01s |
| Harvey's Legal Agent Benchmark | $0.17 | $4.01 | 9m39s | 45m05s |
| Legal Research Bench | $0.25 | $1.53 | 11m42s | 25m04s |
| LegalBench | N/A | N/A | 6.53s | 27.42s |
| EMB | $0.23 | $3.06 | 13m00s | 52m59s |
| Finance Agent (v2) | $0.21 | $1.66 | 6m04s | 16m41s |
| Tax Agent Bench | $0.13 | $0.98 | 5m49s | 10m58s |
| MedCode | N/A | N/A | 35.13s | 2m07s |
| MedScribe | N/A | N/A | 53.83s | 70.94s |
| ProofBench v1.1 | $0.13 | $0.76 | 9m07s | 14m22s |
| BioMysteryBench | $0.23 | $1.23 | 14m50s | 14m18s |
| MysteryMechanism | $0.16 | $1.31 | 8m04s | 27m55s |
| Terminal-Bench Science | $0.47 | $4.82 | 2h57m | 1h18m |
| SAGE | N/A | N/A | 35.78s | 3m22s |
| Code Migration | $0.94 | $14.58 | 1h45m | 1h30m |
| IOI | $0.24 | $8.18 | 18m30s | 3h57m |
| ProgramBench | $0.90 | $16.76 | 1h02m | 4h27m |
| SkillsBench | $0.09 | $0.92 | 3m43s | 13m28s |
| Terminal-Bench 4.0 | $0.50 | $5.07 | 48m22s | 29m27s |
| Vibe Code Bench 1-100 | $0.69 | $26.27 | 28m21s | 2h11m |
| Vibe Code Bench v1.1 | $0.41 | $4.88 | 15m28s | 25m30s |
| CyberBench v1.1 | $0.07 | $4.07 | 13m50s | 36m52s |
| Public Benefits Bench v1.1 | $0.07 | $0.92 | 9m59s | 25m53s |
Results available only for DeepSeek V4.1 Flash
- SRE Bench
Results available only for Grok 4.6
- Vals RSI Index
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- LiveCodeBench
- SWE-bench
- Time Horizon Index: KSP