models
Open weights, fine-tunes and adapters. Every listing here comes live from the Hugging Face Hub, attributed to it, and links back to the source.
Qwen3-4B-Non-Thinking-RL-Math-Step500qwen3-30b-a3b-base-reasoning-sft-nemotron-math-v4-500m-i1-GGUFqwen3-30b-a3b-base-reasoning-sft-nemotron-math-v4-500m-GGUFQwQ-Math-IO-500M-i1-GGUFovd-math-1-data-full-step500-historicalQwQ-Math-IO-500M-GGUFovd-math-1-data-hes-step500-historicalQwen3-4B-OpenThoughts3-Math-SFT-step500LLaDA-Instruct-JustGRPO-Math500ovd-math-1-data-random-step500-historicalQwQ-Math-IO-500M-GGUFDataDecide-dolma1_7-no-math-code-1B-50000_52500_55000_57500_60000_weightedavg_mergeDataDecide-dolma1_7-no-math-code-1B-47500_50000_52500_55000_57500_weightedavg_mergeDataDecide-dolma1_7-no-math-code-1B-45000_47500_50000_52500_55000_weightedavg_mergeDataDecide-dolma1_7-no-math-code-1B-42500_45000_47500_50000_52500_weightedavg_mergellama3_1_8b-instruct-CB_SSFT-lisa_math_rho1.0_step500-500_a16_lr3e-4QwQ-Math-IO-500M-Q4_K_M-GGUFllama3_2_3b-instruct-CB_SSFT-lisa_math_rho1.0_step500-500_a16_lr3e-4SWAP_MATH500_Disc_Llama3-8B-LoRADataDecide-dolma1_7-no-math-code-1B-40000_42500_45000_47500_50000_weightedavg_mergeDataDecide-dolma1_7-no-math-code-1B-0_2500_5000_7500_10000_weightedavg_mergeQwQ-Math-IO-500MDataDecide-dolma1_7-no-math-code-1B-5000_7500_10000_12500_15000_weightedavg_mergeSWAP_v2_MATH500_Gen_Llama3-8B-LoRADataDecide-dolma1_7-no-math-code-1B-2500_5000_7500_10000_12500_weightedavg_mergellama3-bc-math500LightningRL-8B-b32-MATH500SWAP_MATH500_Gen_Llama3-8B-LoRAQwen3-30B-A3B-SFT-Math-Code-1M-500tulu-2-7b_20251112_mixed_tulu3-sft_math500_64_seed1_original_640ex_10_0pct_e1_lr2e_05_bs64