Reward Modeling on RewardBench Math 2
35.7AccuracyPointwise Rating
Evaluation Results
| Method | Links | |
|---|---|---|
| Pointwise RatingModel Family=QWEN3, Model Size=4B2026.02 | 35.7 | |
| Pointwise RatingModel Family=QWEN3, Model Size=8B2026.02 | 35.7 | |
| Listwise Ranking*Model Family=GEMMA3, Model Size=4B2026.02 | 31.8 | |
| FLIPModel Family=QWEN3, Model Size=8B2026.02 | 31 | |
| Listwise Ranking*Model Family=MISTRAL, Model Size=7B2026.02 | 30.3 | |
| Pointwise RatingModel Family=QWEN3, Model Size=1.7B2026.02 | 30.2 | |
| FLIPModel Family=QWEN3, Model Size=1.7B2026.02 | 30.2 | |
| Pairwise Ranking*Model Family=GEMMA3, Model Size=4B2026.02 | 30.2 | |
| Listwise Ranking*Model Family=LLAMA3, Model Size=8B2026.02 | 29.8 | |
| FLIPModel Family=MISTRAL, Model Size=7B2026.02 | 29.7 | |
| FLIPModel Family=QWEN3, Model Size=4B2026.02 | 29.4 | |
| Pairwise Ranking*Model Family=LLAMA3, Model Size=8B2026.02 | 29.3 | |
| FLIPModel Family=OLMO2, Model Size=1B2026.02 | 28.4 | |
| Listwise Ranking*Model Family=QWEN3, Model Size=1.7B2026.02 | 27.2 | |
| FLIPModel Family=GEMMA3, Model Size=4B2026.02 | 27.1 | |
| Pairwise Ranking*Model Family=MISTRAL, Model Size=7B2026.02 | 26.8 | |
| FLIPModel Family=LLAMA3, Model Size=1B2026.02 | 26.6 | |
| FLIPModel Family=OLMO2, Model Size=7B2026.02 | 26.3 | |
| FLIPModel Family=QWEN3, Model Size=0.6B2026.02 | 26.3 | |
| FLIPModel Family=GEMMA3, Model Size=1B2026.02 | 26.2 | |
| FLIPModel Family=LLAMA3, Model Size=8B2026.02 | 25.8 | |
| Listwise Ranking*Model Family=OLMO2, Model Size=1B2026.02 | 24 | |
| FLIPModel Family=LLAMA3, Model Size=3B2026.02 | 23.8 | |
| Pairwise Ranking*Model Family=QWEN3, Model Size=1.7B2026.02 | 23.1 | |
| Listwise Ranking*Model Family=GEMMA3, Model Size=1B2026.02 | 22.8 | |
| Pairwise Ranking*Model Family=OLMO2, Model Size=7B2026.02 | 22.7 | |
| Listwise Ranking*Model Family=LLAMA3, Model Size=3B2026.02 | 22.7 | |
| Pairwise Ranking*Model Family=GEMMA3, Model Size=1B2026.02 | 22.5 | |
| FLIPModel Family=GEMMA3, Model Size=270M2026.02 | 22.4 | |
| Pointwise RatingModel Family=GEMMA3, Model Size=4B2026.02 | 22.1 | |
| Pointwise RatingModel Family=LLAMA3, Model Size=8B2026.02 | 21.4 | |
| Pointwise RatingModel Family=QWEN3, Model Size=0.6B2026.02 | 21.1 | |
| Pointwise RatingModel Family=LLAMA3, Model Size=3B2026.02 | 20.3 | |
| Pointwise RatingModel Family=LLAMA3, Model Size=1B2026.02 | 19.9 | |
| Listwise Ranking*Model Family=OLMO2, Model Size=7B2026.02 | 19 | |
| Pairwise Ranking*Model Family=LLAMA3, Model Size=3B2026.02 | 18.7 | |
| Pointwise RatingModel Family=GEMMA3, Model Size=270M2026.02 | 18.7 | |
| Listwise Ranking*Model Family=LLAMA3, Model Size=1B2026.02 | 17.6 | |
| Pointwise RatingModel Family=GEMMA3, Model Size=1B2026.02 | 16.9 | |
| Pairwise Ranking*Model Family=QWEN3, Model Size=4B2026.02 | 16 | |
| Pointwise RatingModel Family=OLMO2, Model Size=7B2026.02 | 13.2 | |
| Pairwise Ranking*Model Family=QWEN3, Model Size=8B2026.02 | 12.9 | |
| Listwise Ranking*Model Family=GEMMA3, Model Size=270M2026.02 | 11.6 | |
| Pointwise RatingModel Family=OLMO2, Model Size=1B2026.02 | 10.3 | |
| Pointwise RatingModel Family=MISTRAL, Model Size=7B2026.02 | 10.1 | |
| Pairwise Ranking*Model Family=LLAMA3, Model Size=1B2026.02 | 9.7 | |
| Pairwise Ranking*Model Family=QWEN3, Model Size=0.6B2026.02 | 8.2 | |
| Pairwise Ranking*Model Family=GEMMA3, Model Size=270M2026.02 | 5.9 | |
| Listwise Ranking*Model Family=QWEN3, Model Size=0.6B2026.02 | 4.5 | |
| Listwise Ranking*Model Family=QWEN3, Model Size=8B2026.02 | 3.6 | |
| Listwise Ranking*Model Family=QWEN3, Model Size=4B2026.02 | 3.2 | |
| Pairwise Ranking*Model Family=OLMO2, Model Size=1B2026.02 | 1.4 |