Up to four models
Claude Fable 5.2 vs Gemini 4 Argon
On the index (GPT-4 = 100), Claude Fable 5.2 scores 202 and Gemini 4 Argon 191. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Fable 5.2
202185–218
WeightsClosed
ReleasedNov 2026
Price in / out$8.0 / $40
Context1M
Core tests taken4 of 9
191176–207
WeightsClosed
ReleasedSep 2026
Price in / out$2.0 / $10
Context1.05M
Core tests taken6 of 9
Claude Fable 5.2 leads Gemini 4 Argon by 10 points, ahead on 2 of 4 shared core tests, at 4.0× the price.
Test by testthird-party results
Claude Fable 5.2
Gemini 4 Argon
Every result as a table
| Test | Claude Fable 5.2 | Gemini 4 Argon |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 96.0% | 96.4% |
| EBR-bench | 90.5% | 64.8% |
| SimpleBench | 94.9% | 91.7% |
| SimpleQA Verified | 94.0% | 76.0% |
| Maths | ||
| FrontierMath (tiers 1-3) | 95.1% | 86.0% |
| AIME-style maths (OTIS mock) | 99.7% | 100.0% |
| ProofBench v1.1 | 100.0% | 99.0% |
| Chess puzzles | 79.0% | 59.0% |
| FrontierMath (tier 4) | 100.0% | 78.0% |
| Coding | ||
| Terminal-Bench 4.0 | 65.1% | 57.6% |
| ProgramBench | 19.0% | 2.5% |
| Vibe Code Bench v1.1 | 93.8% | 91.9% |
| FrontierSWE | 69.9% | 55.0% |
| Agents | ||
| APEX-Agents | 80.1% | 82.4% |
| BioMysteryBench | 83.7% | 76.3% |
| Terminal-Bench-Science | 55.7% | 44.3% |
| DTBench | 99.2% | 97.6% |
| Furniture assembly | 90.0% | 72.5% |
| LMCA | 70.6% | 63.7% |
| Mystery game puzzles | 90.0% | 64.0% |
| Professional work | ||
| HLAB-Vals-1.1 | 5.0% | 27.1% |
| Legal Research Bench | 46.2% | 54.8% |
| Tax Agent Bench | 32.1% | 49.2% |
| Novel problems | ||
| MysteryMechanism | 51.8% | 45.5% |
Tests at least two of these models have taken. Bold rows are the core tests.