Claude Fable 5.2 vs Gemini 3.8 Flash
Claude Fable 5.2 leads Gemini 3.8 Flash by 20.2 points on the Shogo Agent Index. Claude Fable 5.2 wins 19 and Gemini 3.8 Flash wins 0 of 20 shared benchmarks. Gemini 3.8 Flash is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5.2 | Gemini 3.8 Flash |
|---|---|---|
| Agentic coding | 86 | 63 |
| Long-horizon autonomy | 78 | 64 |
| Knowledge-work agents | 67 | 47 |
| Computer use | 77 | 44 |
| Security & context learning | 65 | 53 |
Shared benchmarks
| Benchmark | Claude Fable 5.2 | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agentsindex | 80.1% | 64.3% |
| Vending-Bench 2index | $9,628 | $5,094 |
| FrontierSWEindex | 69.9% | 19.6% |
| Remote Labor Indexindex | 41.3% | 5.8% |
| CursorBenchindex | 60.0% | 39.6% |
| ARC-AGI | 98.5% | 98.5% |
| ARC-AGI-2 | 93.8% | 89.2% |
| Chess Puzzles | 79.0% | 58.0% |
| CritPt | 35.7% | 18.3% |
| DTBench | 99.2% | 95.7% |
| FrontierMath Tier 4 | 100.0% | 22.0% |
| FrontierMath Tiers 1-3 | 95.1% | 68.4% |
| GPQA Diamond | 96.0% | 95.4% |
| LMCA | 70.6% | 52.9% |
| OTIS Mock AIME | 99.7% | 98.9% |
| ProofBench | 100.0% | 48.0% |
| SciCode | 67.7% | 56.6% |
| SimpleBench | 94.9% | 82.4% |
| SimpleQA Verified | 94.0% | 69.7% |
| WeirdML v3 | 43.1% | 7.4% |
Specs and price
| Claude Fable 5.2 | Gemini 3.8 Flash | |
|---|---|---|
| Agent Index | 76.4 (#1) | 56.2 (#21) |
| Agent $/M | $3.21 | $0.35 |
| Input / output $/M | $8.00 / $40.00 | $0.75 / $3.75 |
| Context | 1M | 1M |
| Released | Nov 9, 2026 | Sep 2, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 6 |
Don’t pick one. Route.
Shogo can run Claude Fable 5.2 and Gemini 3.8 Flash side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo