Fiction.LiveBench - 0 tokens (relevant passages only)
Chat & Writing · 2025-02-19
Fiction.LiveBench quiz accuracy on the 0-token version of each story, cut down to only the passages a question needs. It is the publisher's control condition for the same private comprehension questions asked at every longer length; question counts are not published.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4 | 100.0 |
| Sonnet 4 | 100.0 |
| Sonnet 4.5 | 100.0 |
| DeepSeek-R1-0528 | 100.0 |
| DeepSeek-V3.1 | 100.0 |
| DeepSeek-V3.2-Exp | 100.0 |
| DeepSeek-V3.2 | 100.0 |
| Gemini 2.5 Flash | 100.0 |
| Gemini 2.5 Flash-Lite | 100.0 |
| Gemini 2.5 Pro | 100.0 |
| Gemini 3 Flash Preview | 100.0 |
| Gemini 3 Pro Preview | 100.0 |
| GLM-4.5 | 100.0 |
| GLM-4.7 | 100.0 |
| GLM-5 | 100.0 |
| GPT-4.1 | 100.0 |
| GPT-4.5 | 100.0 |
| GPT-5 | 100.0 |
| GPT-5.2 | 100.0 |
| GPT-5.2 Pro | 100.0 |
| GPT-5 Mini | 100.0 |
| gpt-oss-120b | 100.0 |
| Grok 4 | 100.0 |
| Grok 4 Fast | 100.0 |
| Kimi K2.5 | 100.0 |
| Llama 4 Maverick Instruct | 100.0 |
| MiniMax M2.5 | 100.0 |
| Nemotron 3 Nano 30B A3B | 100.0 |
| o3 | 100.0 |
| O4 Mini | 100.0 |
| Qwen3-235B-A22B | 100.0 |
| Qwen3 235B A22B Thinking 2507 | 100.0 |
| Qwen3 30B A3B Thinking 2507 | 100.0 |
| Qwen3.5 Plus (2026-02-15) | 100.0 |
| Qwen3 8B | 100.0 |
| Qwen3-Next-80B-A3B-Thinking | 100.0 |
| QwQ-32B | 100.0 |
| Sonoma Sky Alpha | 100.0 |
| ChatGPT-4o Latest (source-unspecified snapshot) | 87.5 |
| Opus 4.5 | 87.5 |