DeepSeek V4.1 Flash 1017 vs Gemini 3.8 Flash: Benchmark Comparison
DeepSeek V4.1 Flash 1017 has the higher score on 6 of 9 shared benchmarks; Gemini 3.8 Flash leads on 3.
The largest observed score gap is 12.23 pts on Tax Agent Bench , where DeepSeek V4.1 Flash 1017 leads.
Reported ±1 standard-error ranges overlap on 4 of 9 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 1017 | Gemini 3.8 Flash | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 56.58% ±1.27 | 54.89% ±1.02 | 1.70 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 9.17% ±2.51 | 10.00% ±2.42 | 0.83 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 45.19% ±3.46 | 38.94% ±3.39 | 6.25 pts | Reported ±1 SE ranges overlap |
| EMB | 61.80% ±3.19 | 72.20% ±2.42 | 10.39 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 56.96% ±2.05 | 61.44% ±0.13 | 4.48 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 44.35% ±3.57 | 32.12% ±3.37 | 12.23 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 44.21% ±4.46 | 36.55% ±4.18 | 7.66 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 31.82% ±2.62 | 19.70% ±2.31 | 12.12 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 87.73% ±2.75 | 78.65% ±3.88 | 9.08 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | DeepSeek V4.1 Flash 1017 average | Gemini 3.8 Flash average |
|---|---|---|
| Legal | 27.18% | 24.47% |
| Finance | 54.37% | 55.25% |
| Coding | 54.59% | 44.96% |
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 1017 cost | Gemini 3.8 Flash cost | DeepSeek V4.1 Flash 1017 latency | Gemini 3.8 Flash latency |
|---|---|---|---|---|
| Vals Index | $0.64 | $5.71 | 36m33s | 51m52s |
| Harvey's Legal Agent Benchmark | $0.54 | $3.66 | 21m39s | 29m21s |
| Legal Research Bench | $0.70 | $1.63 | 35m07s | 5m24s |
| EMB | $0.36 | $8.24 | 16m47s | 12m49s |
| Finance Agent (v2) | $0.46 | $2.00 | 14m39s | 3m21s |
| Tax Agent Bench | $0.46 | $0.86 | 21m04s | 2m58s |
| Code Migration | $1.86 | $18.49 | 2h11m | 2h37m |
| Terminal-Bench 4.0 | $0.54 | $8.65 | 41m10s | 1h47m |
| Vibe Code Bench v1.1 | $0.58 | $6.87 | 16m14s | 8m39s |
Results available only for DeepSeek V4.1 Flash 1017
None.
Results available only for Gemini 3.8 Flash
- Vals RSI Index
- Vals Science Index
- LegalBench
- MortgageTax
- TaxEval v2
- MedCode
- MedScribe
- ProofBench v1.1
- BioMysteryBench
- MysteryMechanism
- Terminal-Bench-Science
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- SAGE
- IOI
- LiveCodeBench
- ProgramBench
- SkillsBench
- SWE-bench
- SWE-bench Multimodal
- Vibe Code Bench 1-100
- CyberBench v1.1
- SRE Bench
- Public Benefits Bench v1.1
- CUA-bench
- Time Horizon Index: KSP