Grade-school mathematical reasoning on GSM8K (Throughput)
160.51Output ThroughputOur approach
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Our approachVocabulary size=13264, Hardware=Single A100-80G GPU, Target Model=Llama-3.1-8B-Instruct, Inference Engine=SGLang, Framework=SpecForge2026.03 | 160.51 | 3.2 | |
| 128k-vocabVocabulary size=128k, Hardware=Single A100-80G GPU, Target Model=Llama-3.1-8B-Instruct, Inference Engine=SGLang, Framework=SpecForge2026.03 | 155.47 | — |