Benchmark
MRCR v2 (8-needle)
Long context benchmarks measure how well models use information in long inputs. Scores are percentages.
All long context benchmarksScores from LLM Stats data.
- Models measured
- 26
- Scores recorded
- 26
- Sources
- 1
- IQ weight
- 1.05
Score distribution
- 49.6%
- Mean
- 44.1%
- Median
- 27.3%
- Std dev
- 21.3%
- 10th percentile
- 92.9%
- 90th percentile
- 13.5%
- Min
This benchmark has a weight of 1.05 for chat models, measured over 26 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Muse Spark 1.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 98.5% |
| 2 | Gemini 3.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 97% |
| 3 | Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 92.9% |
| 4 | GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 91.5% |
| 5 | GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.6% |
| 6 | Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 76% |
| 7 | GPT 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 74% |
| 8 | Gemma 4 31B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 66.4% |
| 9 | Gemini 3.1 Flash Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 60.1% |
| 10 | Gemini 3.6 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 54% |
| 11 | Seed 2.0 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 51.4% |
| 12 | Inkling 2 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 46.4% |
| 13 | Gemma 4 26B A4B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 44.1% |
| 14 | Gemma 4 12B It· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43.4% |
| 15 | GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 41.3% |
| 16 | GPT-5.4 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 33.6% |
| 17 | GPT 5.4 Nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 33.1% |
| 18 | Gemini 3.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 26.6% |
| 19 | Gemini 3 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 26.3% |
| 20 | Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 26.3% |
| 21 | Gemma 4 E4B-it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 25.4% |
| 22 | Gemini 3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 22.1% |
| 23 | Gemini 3.5 Flash Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 21.3% |
| 24 | Gemma 4 E2B-it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 19.1% |
| 25 | Gemini 2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 16.4% |
| 26 | Gemma 3 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 13.5% |
Data credits: Data by LLM Stats