Claude Fable 5.1 vs Gemini 4 Argon: Benchmark Comparison
Claude Fable 5.1 has the higher score on 16 of 23 shared benchmarks; Gemini 4 Argon leads on 7.
The largest observed score gap is 21.37 pts on SRE Bench , where Gemini 4 Argon leads.
Reported ±1 standard-error ranges overlap on 13 of 20 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 Fable 5.1 | Gemini 4 Argon | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 36.09% | 30.55% | 5.54 pts | Uncertainty comparison unavailable |
| Legal Research Bench | 55.29% ±3.46 | 54.81% ±3.46 | 0.48 pts | Reported ±1 SE ranges overlap |
| LegalBench | 88.51% ±0.42 | 88.30% ±0.37 | 0.21 pts | Reported ±1 SE ranges overlap |
| EMB v1.1 | 76.73% ±2.13 | 74.83% ±2.49 | 1.90 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 49.19% ±3.61 | 44.56% ±3.59 | 4.63 pts | Reported ±1 SE ranges overlap |
| MedCode | 53.51% ±2.17 | 58.80% ±2.10 | 5.29 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 91.29% ±1.95 | 87.43% ±1.96 | 3.87 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 100.00% ±0.00 | 99.00% ±1.00 | 1.00 pts | Reported ±1 SE ranges overlap |
| MysteryMechanism | 47.75% ±3.36 | 45.49% ±3.35 | 2.25 pts | Reported ±1 SE ranges overlap |
| Robo-bench | 5.83% | 0.83% | 5.00 pts | Uncertainty comparison unavailable |
| Terminal-Bench-Science | 40.00% ±5.90 | 44.29% ±5.98 | 4.29 pts | Reported ±1 SE ranges overlap |
| SAGE | 48.53% ±3.34 | 53.65% ±3.38 | 5.12 pts | Reported ±1 SE ranges overlap |
| IOI | 90.78% ±4.65 | 100.00% ±0.00 | 9.22 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 7.00% ±1.81 | 2.50% ±1.11 | 4.50 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 61.25% ±2.23 | 56.04% ±2.27 | 5.21 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 58.08% ±3.31 | 57.58% ±2.31 | 0.51 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench 1-100 | 28.00% ±4.49 | 21.20% ±3.81 | 6.80 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 90.26% ±1.57 | 91.91% ±1.90 | 1.64 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 70.42% ±5.43 | 77.86% ±5.30 | 7.44 pts | Reported ±1 SE ranges overlap |
| SRE Bench | 22.90% ±2.60 | 44.27% ±3.08 | 21.37 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 74.90% ±1.13 | 69.76% ±0.00 | 5.14 pts | Reported ±1 SE ranges do not overlap |
| CUA-bench | 13.17% | 4.83% | 8.33 pts | Uncertainty comparison unavailable |
| Time Horizon Index: KSP | 63.33% ±0.00 | 52.50% ±8.86 | 10.83 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Fable 5.1 average | Gemini 4 Argon average |
|---|---|---|
| Legal | 71.90% | 71.55% |
| Finance | 62.96% | 59.69% |
| Healthcare | 72.40% | 73.11% |
| Math | 100.00% | 99.00% |
| Science | 31.19% | 30.20% |
| Education | 48.53% | 53.65% |
| Coding | 55.90% | 54.87% |
| Cyber | 46.66% | 61.07% |
| Social Mobility | 74.90% | 69.76% |
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 Fable 5.1 cost | Gemini 4 Argon cost | Claude Fable 5.1 latency | Gemini 4 Argon latency |
|---|---|---|---|---|
| Vals RSI Index | $298.22 | $179.80 | 90h00m | 90h00m |
| Legal Research Bench | $23.06 | $6.45 | 57m29s | 31m21s |
| LegalBench | N/A | N/A | 12.40s | 12.62s |
| EMB v1.1 | $15.93 | $11.80 | 29m32s | 38m26s |
| Tax Agent Bench | $13.18 | $3.47 | 28m34s | 16m26s |
| MedCode | N/A | N/A | 3m35s | 79.53s |
| MedScribe | N/A | N/A | 3m08s | 87.44s |
| ProofBench v1.1 | $2.70 | $0.78 | 7m58s | 6m20s |
| MysteryMechanism | $5.63 | $4.59 | 17m41s | 24m55s |
| Robo-bench | $20.40 | $17.33 | 2m04s | 2m32s |
| Terminal-Bench-Science | $38.01 | $44.90 | 2h36m | 1h46m |
| SAGE | N/A | N/A | 105.75s | 63.07s |
| IOI | $11.20 | $5.31 | 30m06s | 24m19s |
| ProgramBench | $58.52 | $17.66 | 2h27m | 1h11m |
| SWE-bench Multimodal | $5.46 | $11.40 | 15m12s | 33m38s |
| Terminal-Bench 4.0 | $17.18 | $17.64 | 1h00m | 53m35s |
| Vibe Code Bench 1-100 | $149.66 | $18.66 | 2h53m | 1h43m |
| Vibe Code Bench v1.1 | $33.37 | $8.11 | 57m40s | 37m28s |
| CyberBench v1.1 | $4.01 | $5.11 | 18m32s | 19m10s |
| SRE Bench | $32.23 | $30.45 | 2h03m | 1h35m |
| Public Benefits Bench v1.1 | $7.48 | $2.79 | 48m02s | 26m07s |
| CUA-bench | $323.78 | $193.78 | 37.44s | 70.51s |
| Time Horizon Index: KSP | $4655.01 | $708.63 | N/A | N/A |
Results available only for Claude Fable 5.1
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
Results available only for Gemini 4 Argon
- Vals Finance Index
- Vals Index
- Vals Legal Index
- Vals Science Index
- HLAB-Vals-1.1
- Finance Agent v2.1
- BioMysteryBench
- Code Migration v1.1