List your company

Reward Modeling on RewardBench Focus 2

90.3AccuracyRubric-ARM-voting@5
-1.42822.38646.270.014Dec 2, 2025Dec 14, 2025Dec 26, 2025Jan 8, 2026Jan 20, 2026Feb 1, 2026Feb 14, 2026

Evaluation results

MethodLinks
Rubric-ARM-voting@5voting=52026.0290.3
Rubric-ARM2026.0289.4
Claude-3.5-Sonnet2026.0287
Rubric-RM-voting@5voting=52026.0286.5
RM-R1-14B (Qwen-2.5-Inst)Backbone=Qwen-2.5-Inst, Size=14B2026.0284.6
RM-R1-32B (Qwen-2.5-Inst)Backbone=Qwen-2.5-Inst, Size=32B2026.0284.2
Gemini-2.5-Flash2026.0284.1
RRM-32BSize=32B2026.0283.6
Rubric-RM2026.0282.2
API (Rubric+Judge)Rubric Model=GPT-4.1-Mini, Judge Model=Gemini-2.5-Flash Lite2026.0279.6
RM-R1-32B (DeepSeek-Dist)Backbone=DeepSeek-Dist, Size=32B2026.0279.2
RM-R1-14B (DeepSeek-Dist)Backbone=DeepSeek-Dist, Size=14B2026.0279
RM-R1-7B (Qwen-2.5-Inst)Backbone=Qwen-2.5-Inst, Size=7B2026.0276.2
Distribution-Calibrated Aggregationn=12, Judge LLM=gemini-2.5-flash2025.1272.5
FLIPModel Family=QWEN3, Model Size=8B2026.0272.2
FLIPModel Family=MISTRAL, Model Size=7B2026.0269.3
Distribution-Calibrated Aggregationn=4, Judge LLM=gemini-2.5-flash2025.1268.5
FLIPModel Family=QWEN3, Model Size=4B2026.0266.8
Soft-SCn=12, Judge LLM=gemini-2.5-flash2025.1266.3
FLIPModel Family=QWEN3, Model Size=0.6B2026.0265.7
FLIPModel Family=OLMO2, Model Size=7B2026.0265.6
FLIPModel Family=GEMMA3, Model Size=4B2026.0265.5
FLIPModel Family=LLAMA3, Model Size=8B2026.0264.1
Soft-SCn=4, Judge LLM=gemini-2.5-flash2025.1263.6
API (direct Judge)Judge Model=Gemini-2.5-Flash Lite2026.0263.4
SCn=4, Judge LLM=gemini-2.5-flash2025.1262.9
SCn=12, Judge LLM=gemini-2.5-flash2025.1262.6
CI-SCn=4, Judge LLM=gemini-2.5-flash2025.1261.6
CI-SCn=12, Judge LLM=gemini-2.5-flash2025.1261.6
USCn=12, Judge LLM=gemini-2.5-flash2025.1261.2
RRM-7BSize=7B2026.0260.4
USCn=4, Judge LLM=gemini-2.5-flash2025.1260.4
GSCn=12, Judge LLM=gemini-2.5-flash2025.1260.2
GSCn=4, Judge LLM=gemini-2.5-flash2025.1260.1
FLIPModel Family=GEMMA3, Model Size=1B2026.0259.1
FLIPModel Family=QWEN3, Model Size=1.7B2026.0258.6
Qwen-3-8B (Rubric+Judge)Backbone=Qwen-3-8B2026.0256.6
RM-R1-7B (DeepSeek-Dist)Backbone=DeepSeek-Dist, Size=7B2026.0255.4
FLIPModel Family=OLMO2, Model Size=1B2026.0255
Pairwise Ranking*Model Family=QWEN3, Model Size=4B2026.0251.3
FLIPModel Family=GEMMA3, Model Size=270M2026.0250.7
Pairwise Ranking*Model Family=QWEN3, Model Size=8B2026.0248.3
FLIPModel Family=LLAMA3, Model Size=1B2026.0243.3
Listwise Ranking*Model Family=QWEN3, Model Size=4B2026.0242.7
Pointwise RatingModel Family=QWEN3, Model Size=8B2026.0241.3
Pointwise RatingModel Family=QWEN3, Model Size=4B2026.0240.4
Pairwise Ranking*Model Family=QWEN3, Model Size=1.7B2026.0239
Listwise Ranking*Model Family=QWEN3, Model Size=8B2026.0238.5
FLIPModel Family=LLAMA3, Model Size=3B2026.0238.3
Listwise Ranking*Model Family=LLAMA3, Model Size=8B2026.0238.1
Listwise Ranking*Model Family=MISTRAL, Model Size=7B2026.0235.6
Listwise Ranking*Model Family=QWEN3, Model Size=1.7B2026.0235
Listwise Ranking*Model Family=GEMMA3, Model Size=4B2026.0234.7
Pointwise RatingModel Family=QWEN3, Model Size=1.7B2026.0232.4
JudgeLRM-7BSize=7B2026.0229.1
Pairwise Ranking*Model Family=GEMMA3, Model Size=4B2026.0228.4
Pointwise RatingModel Family=GEMMA3, Model Size=4B2026.0226.1
Listwise Ranking*Model Family=OLMO2, Model Size=7B2026.0226
Pairwise Ranking*Model Family=MISTRAL, Model Size=7B2026.0225.7
Pairwise Ranking*Model Family=OLMO2, Model Size=7B2026.0223.2
Listwise Ranking*Model Family=QWEN3, Model Size=0.6B2026.0222.6
Listwise Ranking*Model Family=GEMMA3, Model Size=1B2026.0222
Pairwise Ranking*Model Family=GEMMA3, Model Size=1B2026.0221.8
Listwise Ranking*Model Family=LLAMA3, Model Size=3B2026.0221
Pairwise Ranking*Model Family=LLAMA3, Model Size=8B2026.0220.5
Pairwise Ranking*Model Family=QWEN3, Model Size=0.6B2026.0220.1
Pointwise RatingModel Family=QWEN3, Model Size=0.6B2026.0218.8
Pointwise RatingModel Family=GEMMA3, Model Size=270M2026.0215.3
Pointwise RatingModel Family=LLAMA3, Model Size=8B2026.0214.9
Listwise Ranking*Model Family=OLMO2, Model Size=1B2026.0213.7
Pointwise RatingModel Family=OLMO2, Model Size=1B2026.0213.6
Pairwise Ranking*Model Family=GEMMA3, Model Size=270M2026.0213.4
Listwise Ranking*Model Family=LLAMA3, Model Size=1B2026.0213.2
Pairwise Ranking*Model Family=LLAMA3, Model Size=3B2026.0212.8
Listwise Ranking*Model Family=GEMMA3, Model Size=270M2026.0212.4
Pointwise RatingModel Family=GEMMA3, Model Size=1B2026.0212
Pointwise RatingModel Family=LLAMA3, Model Size=1B2026.0211.5
Pointwise RatingModel Family=LLAMA3, Model Size=3B2026.029.9
Pointwise RatingModel Family=MISTRAL, Model Size=7B2026.029.7
Pairwise Ranking*Model Family=LLAMA3, Model Size=1B2026.027.8
Pointwise RatingModel Family=OLMO2, Model Size=7B2026.027.4
Pairwise Ranking*Model Family=OLMO2, Model Size=1B2026.022.1