Run the fit test →

Catalog / all models

Local LLM catalog.

Models with benchmark scores and a runnable variant.

Check your hardware →

Models listed: 100
Benchmark sources: 2 / Arena + ZeroEval
Formats: 2 / GGUF + SafeTensors

All models Sorted by general score

100 listed

01 Hy3 tencent hy3 92.9 262K 10K 02 Hy4-preview tencent hy4 92.3 1.0M 3K 03 step-3.5-flash AesSedai step 89.8 — 18K 04 minimax-m2.7 AaryanK minimax 89.4 — 2 05 DeepSeek-R1 deepseek-ai DeepSeek 87.3 — 2.8M 06 GLM-5.1 zai-org glm 86.1 203K 8K 07 qwen3-coder-480b-a35b-instruct Qwen qwen-3 85.9 — 94K 08 GLM-4.7 zai-org glm 85.2 — 134K 09 Qwen3.5-397B-A17B Qwen Qwen3.5 83.9 — 1.0M 10 step3-vl-10b seanbailey518 step3-vl 83.9 — 2K 11 Qwen3.5-122B-A10B Qwen Qwen3.5 83.2 — 916K 12 Kimi-K2.6 moonshotai kimi 83.1 — 423 13 DeepSeek-V3.2 deepseek-ai deepseek-v3 82.8 164K 6.1M 14 qwen3-next-80b-a3b-thinking Qwen qwen-3 82.7 — 41K 15 Qwen3.5-27B Qwen Qwen3.5 82.4 — 3.1M 16 GLM-5.3-Flash zai-org glm 81.9 — 34 17 Beam-1.0-501B-A23B reflection beam-1.0 81.5 1.0M 8K 18 glm-4.5v zai-org glm 81.5 — 45K 19 Qwen3.5-9B Qwen qwen-3 81.4 — 6.4M 20 Qwen3.5-35B-A3B Qwen Qwen3.5 81.0 — 3.2M 21 Qwen3.8-27B Qwen qwen-3 80.3 — 7.9M 22 Kimi-K2.5 moonshotai Kimi 80.1 — 6.2M 23 sarvam-30b sarvamai sarvam 79.4 — 56K 24 glm-4.6 zai-org glm 79.4 — 22K 25 Qwen3-Next-80B-A3B-Instruct Qwen qwen-3 79.4 — 530K 26 Qwen3-235B-A22B Qwen qwen-3 78.8 — 627K 27 Qwen3-32B Qwen qwen-3 78.5 — 2.9M 28 Llama-4-Maverick-17B-128E-Instruct meta-llama llama-4-maverick 78.4 — 21K 29 MiniCPM-SALA openbmb minicpm 78.3 524K 1K 30 longcat-flash-chat ggml-org longcat 77.8 — 97 31 Molmo2-8B allenai Molmo2 77.7 — 496K 32 Kimi-K3 moonshotai kimi 77.0 — 862K 33 Athene-V2-Chat Nexusflow athene 76.8 — 1K 34 Qwen3.5-4B Qwen qwen-3 76.2 — 7.2M 35 DeepSeek-V2.5 deepseek-ai deepseek-v2 75.9 164K 7K 36 Llama-3.1-Nemotron-70B-Instruct bartowski llama-3.1 74.8 — 6K 37 qwen3-vl-235b-a22b-instruct Qwen qwen-3 74.5 — 385K 38 QwQ-32B Qwen qwq 74.3 — 70K 39 deepseek-v3.1 unsloth deepseek-v3 74.0 — 3K 40 gemma-2-27b-it google gemma-2 73.7 — 221K 41 glm-4.5 zai-org glm 73.5 — 66K 42 deepseek-r1-zero unsloth deepseek-r1 73.3 — 67 43 Llama-3.1-Tulu-3-70B allenai llama-3.1 73.3 — 1K 44 Qwen3-30B-A3B Qwen Qwen3 73.2 — 1.5M 45 GLM-5.2 zai-org glm 73.0 1.0M 666 46 Phi-4-reasoning-plus microsoft phi-4 72.8 — 5K 47 gemma-2-9b-it-SimPO princeton-nlp gemma-2 72.5 — 456 48 Qwen2.5-Coder-32B-Instruct Qwen qwen-2.5 71.3 33K 513K 49 aya-expanse-32b CohereForAI aya-expanse 70.9 — 0 50 gemma-2-9b-it google gemma-2 70.9 — 269K 51 llama-3.3-nemotron-super-49b-v1 nvidia llama-3.3 70.8 — 32K 52 mimo-v2-flash TeichAI mimo 70.6 — 6K 53 Qwen3-14B Qwen qwen-3 70.4 — 12.8M 54 qvq-72b-preview Qwen qvq 70.3 — 191 55 gemma-4-31B-it google gemma-4 69.9 — 8.3M 56 minimax-m2 unsloth minimax 69.9 — 59K 57 qwen3-vl-32b-thinking Qwen qwen-3 69.3 — 88K 58 gpt-oss-120b openai gpt-oss 69.1 — 2.7M 59 GLM-4.5-Air zai-org glm 69.1 — 402K 60 nemotron-3-super-120b-a12b unsloth nemotron-3-super 68.8 — 116K 61 llama-4-maverick unsloth llama 68.7 — 24K 62 qwen3-235b-a22b-thinking-2507 Qwen qwen-3 67.9 — 81K 63 nvidia-nemotron-nano-9b-v2 nvidia nvidia-nemotron-nano 67.5 — 451K 64 Qwen3-VL-32B-Instruct Qwen Qwen3-VL 67.2 — 1.6M 65 sarvam-105b sarvamai sarvam 67.1 — 18K 66 qwen3-235b-a22b-instruct-2507 Qwen qwen-3 67.1 — 175K 67 Qwen1.5-110B-Chat Qwen qwen1.5 66.8 33K 2K 68 Qwen1.5-72B-Chat Qwen qwen1.5 66.7 33K 10K 69 Mixtral-8x22B-Instruct-v0.1 mistralai mistral 66.2 66K 31K 70 hermes-3-70b mradermacher hermes-3 66.1 — 3K 71 qwen3-vl-4b-thinking Qwen qwen-3 66.0 — 126K 72 gemma-4-26B-A4B-it google gemma-4 65.9 — 6.7M 73 magistral-small-2506 unsloth mistral 65.9 — 2K 74 aya-expanse-8b CohereForAI aya-expanse 65.4 — 0 75 longcat-flash-lite InquiringMinds-AI longcat 65.3 — 1K 76 GLM-4.7-Flash zai-org glm 64.9 — 996K 77 qwen3-vl-8b-thinking Qwen qwen-3 64.6 — 83K 78 phi-4-reasoning microsoft phi-4 64.6 — 16K 79 zephyr-orpo-141b-A35b-v0.1 HuggingFaceH4 zephyr 64.1 — 81 80 Yi-1.5-34B-Chat 01-ai yi-1.5 64.1 4K 13K 81 gpt-oss-20b openai gpt-oss 64.1 — 6.1M 82 pixtral-large mradermacher mistral 64.0 — 81 83 Qwen3.8-Flash-Next Qwen qwen-3 63.8 — 4.0M 84 Qwen3.6-27B Qwen qwen-3 63.7 — 2.3M 85 granite-3.1-8b-instruct ibm-granite granite-3.1 63.6 — 43K 86 Qwen1.5-32B-Chat Qwen qwen1.5 63.0 — 14K 87 gemma-4-E4B-it google gemma-4 62.6 — 321K 88 GLM-5.3 zai-org glm 62.5 1.0M 9K 89 qwen3-vl-235b-a22b-thinking Qwen qwen-3 62.4 — 349K 90 llama-3.1-70b-instruct MaziyarPanahi llama-3.1 62.4 — 103K 91 Phi-3-medium-4k-instruct microsoft phi-3 62.2 — 13K 92 Qwen3.6-35B-A3B Qwen qwen-3 62.0 — 4.6M 93 Qwen1.5-14B-Chat Qwen qwen1.5 61.4 — 20K 94 internlm2_5-20b-chat internlm internlm 61.3 — 745 95 llama-4-scout unsloth llama 61.3 — 46K 96 DeepSeek-V3 deepseek-ai deepseek-v3 60.8 164K 633K 97 qwen2.5-vl-72b lmstudio-community qwen-2.5 60.6 — 3K 98 deepseek-llm-67b-chat deepseek-ai deepseek-llm 60.6 — 3K 99 Yi-34B-Chat 01-ai yi 60.5 4K 28K 100 granite-3.0-8b-instruct ibm-granite granite-3.0 60.4 — 25K

Fit test

Don't read 100 rows.

Run the fit test →