Benchmarks
347 sources · 1,258 models × 3,038 benchmarks · 158,912 scoresFiction.LiveBench, published by Fiction.live, tests deep comprehension of long fiction: models answer private, open-ended quiz questions about character motivations, changing relationships, secrets and foreshadowing in stories cut to lengths from a relevant-passages-only version (0 tokens) up to 192K tokens. This suite-level row defines the family and has no direct score observations; Atlas records each context length on its own child benchmark.
No signal or difficulty estimate. Held out of the fit by design.
This is a benchmark-family page. Scores are recorded on the component benchmarks below rather than directly on the parent, because the components use different score scales and directions.
| Metric | Models | Leader | Direction | Date |
|---|---|---|---|---|
| Fiction.LiveBench - 0 tokens (relevant passages only) | 54 | Opus 4100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 16K tokens | 54 | Opus 4.6100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 1K tokens | 54 | Opus 4.5100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 2K tokens | 54 | Gemini 3 Flash Preview100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 400 tokens | 54 | Opus 4.5100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 4K tokens | 54 | Sonnet 4100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 8K tokens | 54 | Opus 4.6100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 32K tokens | 50 | Gemini 3 Flash Preview100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 60K tokens | 50 | Gemini 3 Pro Preview100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 120K tokens | 43 | Gemini 3 Flash Preview100.0 | higher is better | 2025-09-29 |
| Fiction.LiveBench - 192K tokens | 25 | Gemini 3 Flash Preview100.0 | higher is better | 2025-09-29 |