Long-context retrieval on RULER-NIAH 16k
100AccuracyBF16
Evaluation Results
| Method | Links | |
|---|---|---|
| BF16Backbone=GLM-4.7-FP8†2026.05 | 100 | |
| QuaRot-INT2Backbone=GLM-4.7-FP8†2026.05 | 100 | |
| OSCARBackbone=GLM-4.7-FP8†2026.05 | 100 | |
| BF16Backbone=Qwen3-4B-Thinking-25072026.05 | 99.7 | |
| BF16Backbone=Qwen3-8B2026.05 | 98.9 | |
| OSCARBackbone=Qwen3-4B-Thinking-25072026.05 | 97.8 | |
| OSCARBackbone=Qwen3-8B2026.05 | 93.9 | |
| QuaRot-INT2Backbone=Qwen3-8B2026.05 | 19 | |
| QuaRot-INT2Backbone=Qwen3-4B-Thinking-25072026.05 | 0 |