Claude Fable 5.2 Tops Signal65 Coding Tests at 28/30, Costs $30.51 Per Passed Job
ryanshrout · x · 2026-11-12
Benchmark firm Signal65's PINNACLE board shows Claude Fable 5.2 is the best model it has tested for both enterprise agent work and coding — and on coding, "it's not close."
- Passed 28 of 30 Frontier Class coding jobs; no other model passed more than 21
- Also the most expensive: $30.51 per passed job, about 3.9x Claude Opus 5.5
- Verdict: best on the board, priced like it
Ryan Shrout adds it's "the best coding model we have tested and it is not close."
More from Models
- SubQ Snap 1.2 launches, ties for #1 on Artificial Analysis Decision Index at 73 — ryanallred · 2026-11-12
- inkling 2 shows strong alignment on honeybench, second only to frontier models on cheating — ChowdhuryNeil · 2026-11-12
- Gemini 4 Carbon tops Blueprint-Bench 2 physical-world benchmark, Claude Fable 5.2 second — cgarciae88 · 2026-11-12
- "Idle compaction by default is stochastic lobotomy": a one-line takedown of auto context compaction — generativist · 2026-11-12
- Fable 5.2 can build production software unattended overnight—but Opus still wins for writing — every · 2026-11-12
- Microsoft's MAI Transcribe 2 sees 12x usage growth, hitting 8.5M weekly transcriptions — mustafasuleyman · 2026-11-12