What's new
One entry for each run whose results are on the leaderboard now, newest first. A model that was tested again appears under the run its current result comes from. Models withheld from the ranking are counted, not listed. Also available as an Atom feed.
2026-11-14
9 models on the board from this run, 9 ranked; 2 models withheld from the ranking.
DeepSeek-R1-0528-Qwen3-8B, DeepSeek-R1-Distill-Qwen-14B, DeepSeek-R1-Distill-Qwen-32B, DeepSeek-R1-Distill-Qwen-7B, gpt-oss-20b, Magistral Small (2506), Olmo-3-7B-Think, Qwen3-4B-Thinking-2507, QwQ 32B
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, deepseek/deepseek-v4.1-flash, openai/gpt-5.6-luna. Run r20261105-bv1-2abe78
2026-11-05
33 models on the board from this run, 33 ranked.
Aya Expanse 32B, Codellama 34B Instruct, Codestral 22B V0.1, Command R 35B 08 2024, Deepseek Coder 33B Instruct, Devstral 24B Small 2505, Exaone3.5 32B Instruct, Flint 40B A2B, Gemma2 27B Instruct, Granite4 Small H, Granite4.1 30B, Granite4.2 30B, Granite4.2 3B, Granite4.2 8B, Laguna XS 2.1, Lfm2 24B, Lfm3 1.7B, Lfm3 VL 2B, Medgemma 27B IT, Mistral Small 22B Instruct 2409, Mistral Small 24B Instruct 2501, Mistral Small3.1 24B Instruct 2503, Mixtral 8x7B Instruct V0.1, Muse Glimmer 30B, Nanbeige_Nanbeige4.3-33B, Nemotron 3.5 Bolt E4B, Nemotron 3.5 Lightning 30B A3B, Nemotron Cascade 2 30B A3B, Nemotron3 33B, Olmo 3.1 32B Instruct, Sage-A1.5-Q4_K_M, Spark-X2.6-4B, YI 34B Chat V1.5
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261102-bv1-5246ab
2026-10-29
1 model on the board from this run, 1 ranked.
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261029-bv1-9ddd5f
2026-10-29
1 model on the board from this run, 1 ranked.
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261029-bv1-3e4075
2026-10-28
2 models on the board from this run, 2 ranked.
Qwen2 Math 1.5B Instruct, Qwen2 Math 7B Instruct
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261028-bv1-ef0895
2026-10-28
1 model on the board from this run, 1 ranked.
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261028-bv1-2e24cd
2026-10-28
4 models on the board from this run, 4 ranked.
Falcon3 10B Instruct, Falcon3 1B Instruct, Falcon3 3B Instruct, Falcon3 7B Instruct
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261028-bv1-7ce712
2026-10-26
102 models on the board from this run, 101 ranked; 7 models withheld from the ranking.
Aya 8B 23, Aya Expanse 8B, Codegeex4 9B All, Codegemma 2B Code V1.1, Codegemma 7B Instruct V1.1, Codeqwen 7B Chat V1.5, Cogito 14B V1 Preview Qwen, Cogito 3B V1 Preview Llama, Cogito 8B V1 Preview Llama, Command R7B Arabic 7B 02 2025, Deepcoder 1.5B Preview, Deepcoder 14B Preview, Deepscaler 1.5B Preview, Deepseek Coder 1.3B Instruct, Deepseek Llm 7B Chat, Deepseek R1 1.5B Qwen Distill, Deepseek R1 8B Llama Distill, Deepseek V2 16B Lite Chat, Dolphin Phi 2.7B V2.6, Exaone Deep 2.4B, Exaone Deep 7.8B, Exaone3.5 2.4B Instruct, Falcon2 11B, Gemma 2B Instruct V1.1, Gemma 7B Instruct V1.1, Gemma2 2B Instruct, Gemma3 1B IT, Gemma3n E2B IT, Gemma4 E2B IT, Gemma4 E4B IT, Glm4 9B Chat, Granite Code 3B Instruct, Granite Code 8B Instruct, Granite3 Dense 2B Instruct, Granite3 Dense 8B Instruct, Granite3 Moe 1B Instruct, Granite3 Moe 3B Instruct, Granite3.1 Dense 2B Instruct, Granite3.1 Dense 8B Instruct, Granite3.1 Moe 1B Instruct, Granite3.1 Moe 3B Instruct, Granite3.2 2B Instruct, Granite3.2 8B Instruct, Granite3.3 2B, Granite3.3 8B, Granite4 1B H, Granite4 350M H, Granite4 3B, Granite4 Micro H, Granite4 Tiny H, Granite4.1 8B, Hermes3 3B Llama3.2, Internlm2 7B Chat V2.5, Lfm2.5 8B A1B, Lfm2.5 Thinking 1.2B, Llama3.2 1B Instruct, Marco O1 7B, Mathstral 7B V0.1, Ministral 3 3B Instruct 2512, Ministral 3 8B Instruct 2512, Mistral 7B Instruct V0.3, Mistral Nemo 12B Instruct 2407, Nemotron Mini 4B Instruct, Olmo2 7B 1124 Instruct, Opencoder 1.5B Instruct, Openthinker 7B V2, Ornith 9B, Phi 2.7B Chat V2, Phi3 14B Medium 128K Instruct, Phi3 3.8B Mini 128K Instruct, Phi4 Mini Reasoning 3.8B, Qwen 0.5B Chat V1.5, Qwen 1.8B Chat V1.5, Qwen 4B Chat V1.5, Qwen2 0.5B Instruct, Qwen2 1.5B Instruct, Qwen2.5 0.5B Instruct, Qwen2.5 1.5B Instruct, Qwen2.5 Coder 0.5B Instruct, Qwen2.5 Coder 1.5B Instruct, Qwen3 0.6B, Qwen3 1.7B, Qwen3 VL 2B Instruct, Qwen3 VL 2B Thinking, Qwen3.5 2B, Sailor2 1B Chat, Sailor2 8B Chat, Smallthinker 3B Preview, Smollm2 1.7B Instruct, Smollm2 135M Instruct, Smollm2 360M Instruct, Solar 10.7B Instruct V1, Stable Code 3B Instruct, Stablelm Zephyr 3B, Stablelm2 1.6B Chat, Stablelm2 12B Chat, Translategemma 12B IT, Tulu3 8B, YI 6B Chat V1.5, YI 9B Chat V1.5, YI Coder 1.5B Chat, YI Coder 9B Chat
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261016-bv1-473145
2026-10-23
7 models on the board from this run, 7 ranked.
Codellama 13B Instruct, Codellama 7B Instruct, Deepseek Coder 6.7B Instruct, Llama2 13B Chat, Olmo2 13B 1124 Instruct, Qwen 14B Chat V1.5, Qwen 7B Chat V1.5
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261022-bv1-54a19b
2026-10-15
1 model on the board from this run, 1 ranked.
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261015-bv1-5ac997
2026-10-14
10 models on the board from this run, 10 ranked.
gpt-oss-120b, Llama 4 Scout 17B-16E Instruct, Llama-3.3-70B-Instruct, Qwen3-Coder-Next, Qwen3-VL-30B-A3B-Instruct, Qwen3-VL-32B-Instruct, Qwen3.5-35B-A3B, Qwen3.6-27B, Qwen3.6-35B-A3B, Qwen3.8-27B
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, deepseek/deepseek-v4.1-flash, openai/gpt-5.6-luna. Run r20261012-bv1-6a1291
2026-10-11
57 models on the board from this run, 57 ranked; 1 model withheld from the ranking.
Command R7B (12-2024), DeepSeek-Coder-V2-Lite-Instruct, Devstral Small 2 24B Instruct (2512), dolphin-2.9.1-yi-1.5-34b, EXAONE-3.5-7.8B-Instruct, Gemma 3n E4B IT, Gemma 4 12B IT, gemma-2-9b-it, gemma-3-12b-it, gemma-3-27b-it, gemma-3-4b-it, gemma-4-26B-A4B-it, gemma-4-31B-it, GLM-4.7-Flash, granite-4.1-3b, Hermes-3-Llama-3.1-8B, Llama-2-7b-chat-hf, Llama-3.1-8B-Instruct, Llama-3.2-3B-Instruct, medgemma-4b-it, Meta-Llama-3-8B-Instruct, Ministral 3 14B Instruct (2512), Mistral Small 3.2 24B Instruct (2506), Mistral-7B-Instruct-v0.2, NVIDIA Nemotron 3 Nano 30B-A3B, Olmo-3-7B-Instruct, Ornith-1.5-9B, Phi-3-mini-4k-instruct, Phi-3.5-mini-instruct, phi-4, Phi-4-mini-instruct, Qwen2-VL-7B-Instruct, Qwen2.5-14B-Instruct, Qwen2.5-32B-Instruct, Qwen2.5-3B-Instruct, Qwen2.5-7B-Instruct, Qwen2.5-Coder-14B-Instruct, Qwen2.5-Coder-32B-Instruct, Qwen2.5-Coder-3B-Instruct, Qwen2.5-Coder-7B-Instruct, Qwen2.5-VL-32B-Instruct, Qwen2.5-VL-3B-Instruct, Qwen2.5-VL-7B-Instruct, Qwen3-14B, Qwen3-30B-A3B, Qwen3-30B-A3B-Instruct-2507, Qwen3-32B, Qwen3-4B, Qwen3-4B-Instruct-2507, Qwen3-8B, Qwen3-Coder-30B-A3B-Instruct, Qwen3-VL-4B-Instruct, Qwen3-VL-8B-Instruct, Qwen3.5-27B, Qwen3.5-4B, Qwen3.5-9B, SmolLM3-3B
Battery v1; judge rubric v1; judges anthropic/claude-haiku-5.5, openai/gpt-5.6-luna. Run r20261006-bv1-53e146