diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..9f2e848 --- /dev/null +++ b/.gitignore @@ -0,0 +1,2 @@ +.DS_Store +__pycache__/ diff --git a/__pycache__/main.cpython-312.pyc b/__pycache__/main.cpython-312.pyc deleted file mode 100644 index ed938d3..0000000 Binary files a/__pycache__/main.cpython-312.pyc and /dev/null differ diff --git a/main.py b/main.py index d6ced59..0bba437 100644 --- a/main.py +++ b/main.py @@ -36,231 +36,534 @@ HTTP_PORT = 8080 # 模型列表 # ══════════════════════════════════════════════════════════ ALL_MODEL_IDS = [ - "Alienpenguin10/M3PO-bahdanau-trial1-seed123", - "sujalrajpoot/TrueSyncAI-Aurion", - "prithivMLmods/Tureis-Qwen3_QWQ-4B-Exp", - "standrey/listing-parser-llama31-8b-ft-v1-full", - "zarakiquemparte/zarablend-l2-7b", - "linzju/Bio-Medical-Llama-3-8B_EnchTable_FFN", - "leonMW/Qwen3-4B-Thinking-2507-GSPO-Easy", - "longvideoagent/longvideoagent-qwen3-4b", - "ishikaa/acquisition_qwen3b_alpaca_proximity", - "01ai/Yi-9B-200K", - "Gille/StrangeMerges_33-7B-slerp", - "sstoica12/acquisition_llama-3_2-3b_bins_medmcqa_gradient", - "speechlessai/speechless-coding-7b-16k-tora", - "unsloth/Qwen2.5-Math-1.5B-Instruct", - "Yuma42/KangalKhan-Sapphire-7B", - "shadowml/BeagleSempra-7B", - "bralynn/test18", - "m-a-p/OProver-8B-Round1", - "yeen214/test_llama2_7b", - "Xwin-LM/Xwin-LM-7B-V0.2", - "FreedomIntelligence/AceGPT-13B", - "Edcastro/tinyllama-edcastr_JavaScript-v2", - "zarakiquemparte/zaraxe-l2-7b", - "MaziyarPanahi/Llama-3-8B-Instruct-v0.8", - "defog/sqlcoder2", - "SawinuCP/bus_booking_voice_agent_merged", - "j05hr3d/Llama-3.2-3B-Instruct-C_M_T-DOLLY-SEED999", - "Changgil/K2S3-SOLAR-11b-v1.0", - "agentica-org/DeepCoder-1.5B-Preview", - "prithivMLmods/Omni-Reasoner3-Merged", - "ibm-granite/granite-7b-instruct", - "zarakiquemparte/kuchiki-1.1-l2-7b", - "MaziyarPanahi/Llama-3-8B-Instruct-v0.1", - "Magpie-Align/Llama-3.1-8B-Magpie-Align-v0.1", - "prithivMLmods/Neumind-Math-7B-Instruct", - "dphn/dolphin-2.9.3-qwen2-1.5b", - "OpenBuddy/openbuddy-mistral-7b-v13", - "kairawal/Qwen3-4B-TL-SynthDolly-1A-E3", - "sahilnagaralu/movie-script", - "xw1234gan/SFT_Qwen2.5-1.5B-Instruct_cnk12", - "yunjae-won/ubq30i_qwen4b_sft_yl", - "1010happy/qwen3BInstruct_ClaudeDefault", - "yunjae-won/ubq30i_qwen4b_sft_both", - "willieseun/AIMO-Qwen2.5-Math-1.5B-Instruct-Finetuned", - "NousResearch/Yarn-Mistral-7b-64k", - "xiaolesu/OsmosisProofling-GRPO-NT", - "health360/Healix-410M", - "RUC-AIBOX/STILL-3-1.5B-preview", - "cjvt/GaMS-1B", - "Lansechen/Qwen2.5-7B-Open-R1-GRPO-math-lighteval-1epochstop-withformat", - "Charlie911/vicuna-7b-v1.5-general-temporal-merged", - "Kyleyee/cDPO_hh-seed5", - "prithivMLmods/Llama-3.2-3B-Math-Oct", - "Kyleyee/HINGE_hh-seed5", - "arcee-ai/Patent-Instruct-7b", - "jb723/cross_lingual_epoch2", - "davidkim205/komt-mistral-7b-v1", - "kalisai/Nusantara-1.8b-Indo-Chat", - "prithivMLmods/Llama-8B-Distill-CoT", - "Kyleyee/HINGE_hh-seed3", - "simplescaling/s1.1-1.5B", - "sail/Sailor2-3B-SFT", - "allenai/OLMoE-1B-7B-0924-Instruct", - "prithivMLmods/Llama-3.2-6B-AlgoCode", - "CloneBO/OracleLM", - "NousResearch/Yarn-Solar-10b-32k", - "m-a-p/OProver-8B-Base", - "HuggingFaceH4/mistral-7b-sft-alpha", - "Hyeongwon/P2-split2_prob_Qwen3-4B-Base_0312-01", - "martyn/mixtral-megamerge-dare-8x7b-v1", - "Kyleyee/rDPO_hh-seed4", - "jingyeom/seal3.1.6n_7b", - "sonthenguyen/OpenHermes-2.5-Mistral-7B-mt-bench-DPO-reversed_corrupted", - "shibing624/ziya-llama-13b-medical-merged", - "datajuicer/LLaMA-1B-dj-refine-150B", - "ajibawa-2023/Uncensored-Jordan-7B", - "nlpguy/AlloyIngot", - "HuggingFaceTB/SmolLM3-3B", - "allenai/codetulu-2-7b", - "vihangd/dopeyshearedplats-2.7b-v1", - "uukuguy/speechless-code-mistral-7b-v2.0", - "j05hr3d/Llama-3.2-3B-Instruct-C_M_T-ALPACA", - "cognitivetech/Mistral-7B-Inst-0.2-Bulleted-Notes", - "sail/Qwen2.5-Math-1.5B-Oat-Zero", - "reaperdoesntknow/SMOLM2Prover", - "Kyleyee/ORPO_hh-seed5", - "CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct", - "NousResearch/Nous-Capybara-7B-V1", - "Vijay3548/InterviewMaster-Llama3.1", - "ReviewHub/qwen3-4b-it-2507-sft-2018-2022-rl-step-20", - "Nos-PT/Llama-Carvalho-PT", - "Gille/StrangeMerges_49-7B-dare_ties", - "Kyleyee/CPO_hh-seed2", - "Kyleyee/cDPO_hh-seed3", - "Kyleyee/DrDPO_hh-seed2", - "sthenno-com/miscii-14b-0218", - "mtgv/MobileLLaMA-2.7B-Chat", - "Novaciano/Alice_In_The_Dark_2-Slerp-RP-3.2-1B", - "Himitsui/KuroMitsu-11B", - "owlninjam/nytheria-3b", - "Kyleyee/DrDPO_hh-seed4", - "Kyleyee/DrDPO_hh-seed5", - "shahzebnaveed/NeuralHermes-2.5-Mistral-7B", - "plaguss/mistal-7b-prm-openrlhf", - "viethq188/Rabbit-7B-v2-DPO-Chat", - "EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984", - "dbpedia/nspm-starcoder-1b", - "psh3333/llama-3.2-3b-grpo-merged", - "saarvajanik/facebook-opt-6.7b-qcqa-ub-16-best-for-KV-cache", - "EmbeddedLLM/Mistral-7B-Merge-14-v0.3", - "xformAI/facebook-opt-125m-qcqa-ub-6-best-for-KV-cache", - "Rev124/llama-3-pruned", - "RatanRohith/NeuralPizza-7B-V0.3", - "mncai/DPO_BC_partial_epoch6", - "zeemen2723/museai-lyrics-gen", - "automerger/OgnoExperiment27-7B", - "sohamb37lexsi/qwen25-3b-legal-correction", - "swift/Meta-Llama-3-8B", - "m-a-p/MuPT-v0-8192-190M", - "Kquant03/Samlagast-7B-laser-bf16", - "VTSNLP/Llama3-ViettelSolutions-8B", - "vihangd/dopeyshearedplats-1.3b-v1", - "HCY123902/qwen25_7b_base_hc_tsss_n32_r1_dpo", - "coder3101/gemma-3-1b-it-heretic", - "Hemkant04/qwen05-resume-job-match-evaluator", - "yekon9/Qwen3-4B-Instruct-2507-heretic", - "uukuguy/Orca-2-13b-f16", - "Kquant03/NeuralTrix-7B-dpo-relaser", - "sambanovasystems/SambaLingo-Russian-Base", - "bineric/NorskGPT-Mistral-7b", - "ReviewHub/qwen3-4b-it-2507-sft-2018-2022-rl-step-10", - "beyoru/EvolLLM", - "driaforall/Dria-Agent-a-3B", - "tlphams/zoyllm-7b-slimorca", - "QuixiAI/WizardLM-33B-V1.0-Uncensored", - "bilalRahib/TinyLLama-NSFW-Chatbot", - "Ahatsham/Llama-3-8B-Instruct_Planning_Feedback_oldaug_v2", - "unsloth/Qwen2.5-Math-1.5B", - "wang7776/Llama-2-7b-chat-hf-30-sparsity", - "hector-gr/RLCR-v4-ks-uniqueness-buf5k-hotpot", - "m-a-p/neo_7b_instruct_v0.1", - "p208p2002/llama-traditional-chinese-120M", - "tartuNLP/Llammas-base", - "riotu-lab/ArabianGPT-01B", - "openaccess-ai-collective/DPOpenHermes-7B", - "Enxin/MovieChat-vicuna", - "Skywork/Skywork-OR1-Math-7B", - "Lugha-Llama/Lugha-Llama-8B-wura_edu", - "laion/allenai-sera-unified-316__Qwen3-8B", - "T1anyu/DeepInnovator", - "mideind/icelandic-gpt-sw3-6.7b-gec", - "Rayeeennnnnnnn/legalmind-chatbot", - "Wanfq/FuseLLM-7B", - "mrvinph/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-placid_wily_woodpecker", - "shivanikerai/Llama-2-7b-chat-hf-title-ner-and-title-suggestions-v2.0", - "Invalid-Null/PeiYangMe-0.7", - "open-unlearning/unlearn_tofu_Llama-3.2-1B-Instruct_forget10_GradDiff_lr1e-05_alpha5_epoch5", - "Tesslate/Tessa-T1-3B", - "LorenaYannnnn/unsafe_compliance-Qwen3-0.6B-OURS_self-seed_1", - "HiTZ/latxa-7b-v1", - "huggyllama/llama-7b", - "FlyPig23/Llama3.2-3B_Paper_Impact_media_SFT_1ep", - "hongzhouyu/FineMedLM-o1", - "shisa-ai/shisa-v1-llama3-8b.lr-5e6", - "senseable/Westlake-7B", - "simplescaling/s1.1-3B", - "Borjan/finki-gpt-140M", - "TinyLlama/TinyLlama_v1.1_chinese", - "Manirajan/interview_tiny", - "ajibawa-2023/Young-Children-Storyteller-Mistral-7B", - "TinyLlama/TinyLlama_v1.1_math_code", - "manotham/Thai-dialogue-transalate_sft_80K", - "nassimjp/Maral-7B-alpha-1", - "MSL7/INEX16-7b", - "Rayeeennnnnnnn/mizan-legal-tunisian", - "maheshrawat18/Qwen3-4B-2507-sft1", - "prithivMLmods/Theta-Crucis-0.6B-Turbo1", - "mlabonne/llama-2-7b-miniguanaco", - "FreekCoolAI/privacy-gemma-qlora", - "lex-hue/Delexa-V0.1-7b", - "goldfish-models/tur_latn_100mb", - "shisa-ai/ablation-18-rafbestseq-shisa-v2-llama-3.1-8b-lr8e6", - "yash-lulla/Legal_AI_Assistant", - "prithivMLmods/Deepthink-Llama-3-8B-Preview", - "prithivMLmods/QwQ-R1-Distill-7B-CoT", - "sstoica12/acquisition_metamath_llama_instruct-3_1-8b-math_format_500_combined_openr1math", - "goldfish-models/rus_cyrl_1000mb", - "goldfish-models/ukr_cyrl_1000mb", - "eren23/dpo-binarized-NeutrixOmnibe-7B", - "RAANA-IA/Gheya-med", - "mesolitica/malaysian-tinyllama-1.1b-16k-instructions-v2", - "bue0912/ToolOmni-Qwen3-4B", - "BRlkl/distill-sft-qwen3-4b-full", - "prithivMLmods/PocketThinker-QwQ-3B-Instruct", - "prithivMLmods/Megatron-Bots-1.7B-Reasoning", - "kumarprince070107/geobot", - "ClaudioSavelli/FAME_GA_llama32-1b-instruct-qa", - "kairawal/Qwen3-0.6B-EL-SynthDolly-1A-E8", - "prithivMLmods/Poseidon-Reasoning-1.7B", - "PKU-Alignment/ProgressGym-HistLlama3-8B-C014-instruct-v0.2", - "unsloth/Qwen2-7B", - "prithivMLmods/Open-Xi-Math-Preview", - "MiniLLM/teacher-gpt2-1.5B", - "Rakancorle1/qwen2.5-7b_Instruct_policy_traj_30k_full", - "prithivMLmods/Lang-Exster-0.5B-Instruct", - "prithivMLmods/Nenque-MoT-0.6B-Elite14", - "posicube/Llama2-chat-AYT-13B", - "lomahony/pythia-70m-helpful-sft", - "kyubeen/code-grpo-checkpoint-950", - "nicholasKluge/TeenyTinyLlama-160m", - "allenai/Llama-3.1-Tulu-3-8B", - "BarraHome/Mistroll-7B-v2.2", - "posicube/Llama-chat-AY-13B", - "hongli-zhan/MINT-empathy-Qwen3-4B", - "jhhj25/qwen3-moe-neuron_structure_drop-p50-s1k-128samples-sft", - "Ikonz-Studios/seva-sarathi-intent-qwen3-1.7b", - "NECOUDBFM/Jellyfish-8B", - "gradients-io-tournaments/augmented-1db17e1d682d23fd", - "Locutusque/LocutusqueXFelladrin-TinyMistral248M-Instruct", - "ChuGyouk/R16", - "oveja1122/toolcalling-merged-demo", - "kmseong/llama3_2_3b-instruct-math-safedelta-scale0.8", - "EleutherAI/SmolLM2-1.7B-magpie-ultra-v1.0-math-431k-s", - "paulml/NeuralOmniBeagleMBX-v3-7B", - "skemessage/Qwen2.5-7B-Instruct-neuron", + "Hyeji0101/qwen2_5_1_5b_demo", + "GenueAI/geode-onyx", + "GM77/qwen3-4b-verilog-grpo", + "ChuGyouk/F_R13_T2", + "ChuGyouk/R17", + "Ingingdo/bit-0.5b-final-logic", + "beomi/Llama-3-Open-Ko-8B", + "Fiscus/trinitite_safe_rl_base_model", + "ChuGyouk/F_R12_T3", + "ChuGyouk/F_R12_T2", + "xw1234gan/cnk12_Main_fixed_SFTanchor_3B_step_9", + "xw1234gan/cnk12_Main_fixed_SFTanchor_3B_step_10", + "xw1234gan/cnk12_Main_fixed_BaseAnchor_3B_step_1", + "kmseong/llama3_2_3b-instruct-math-safedelta-scale0.99", + "opencompass/anah-v2", + "ChuGyouk/R14", + "trishajean/qwen-math-cebuano-1.5b-merged", + "GyanAISystems/Gyan-AI-G1-Official", + "Divij/Qwen2.5-3B-Instruct-sft-without-thoughts", + "Divij/Qwen2.5-3B-Instruct-sft-with-thoughts", + "ChuGyouk/R5_1", + "ChuGyouk/R18_1", + "ChuGyouk/R19_1", + "ChuGyouk/R12", + "ChuGyouk/F_R11_T4", + "ChuGyouk/F_R12", + "ChuGyouk/F_R11_T2", + "ChuGyouk/F_R11_T3", + "ChuGyouk/F_R13_1_T1", + "ChuGyouk/F_R12_T4", + "automerger/T3qm7xNeuralsirkrishna-7B", + "Ford91/clifford-ai-v2", + "ChuGyouk/R16_1", + "ChuGyouk/R15_1", + "nkatara/gita-text-generation-gpt2", + "HINT-lab/Qwen2.5-7B-Instruct-Self-Calibration", + "thirdeyeai/Qwen2.5-1.5B-Instruct-uncensored", + "karaselerm/qwen2.5-1.5b-instruct-ru-abliterated-hw6", + "xw1234gan/cnk12_Main_fixed_BaseAnchor_3B_step_2", + "ontocord/wide_3b_sft_stage1.1-ss1-with_intr_math.no_issue", + "mncai/Foundation_Law_epoch4", + "gauri0508/med-record-audit-qwen2.5-3b-grpo", + "unsloth/Phi-4-mini-instruct", + "E-motionAssistant/qwen-2.5-3b-tamil-therapy-merged", + "EscapeJeju/qwen2_5_1_5b_demo", + "AgPerry/Qwen3-8B-fim-v2v3pt-swe-lego-posttrain", + "ChuGyouk/F_R11", + "ChuGyouk/F_R11_1_T1", + "LorenaYannnnn/general_reward-Qwen3-0.6B-OURS_self-seed_1", + "Vortex5/Crimson-Constellation-12B", + "cloudyu/mistral_11B_instruct_v0.1", + "pkupie/Qwen2.5-3B-ug-cpt", + "iproskurina/qwen-hf-fewshot-iter-np-iter3", + "ontocord/wide_3b_sft_stage1.2-ss1-expert_wiki", + "kmseong/llama3_2_3b-instruct-math-safedelta-scale2", + "Thrillcrazyer/Qwen-2.5-1.5B_TAC_Teacher_Qwen32B", + "nyu-dice-lab/VeriThoughts-Reasoning-7B", + "ontocord/wide_3b", + "silvercoder67/Mistral-7b-instruct-v0.2-summ-sft-e2m", + "lihaoxin2020/qwen3-4b-sft-gpt54-ep2-instance-rubric-gpt54-step200", + "lihaoxin2020/qwen3-4b-sft-gpt54-ep2-instance-rubric-gpt54-step150", + "lihaoxin2020/qwen3-4b-sft-gpt54-ep2-evolving-rubric-gem3-flash-step150", + "Guilherme34/Firefly-V3", + "EphAsad/Mnemosyne-3B", + "Sangsang/ci_feedback_both_feedback_jsd_b0p8", + "ChuGyouk/F_R12_1", + "AgPerry/SWE-Lego-Qwen3-4B-posttrain", + "5CH5/Qwen2.5-7B-abliterated", + "55mvresearch/Qwen2.5-7B-Instruct-SFT-FT1-Merged", + "dadaguai6677/TourismReview-Qwen2.5-7B", + "Ramikan-BR/Qwen2-0.5B-v25", + "Anonymous-2004/asgn2-sft_resta", + "Anonymous-2004/asgn2-model_sft_resta", + "xw1234gan/cnk12_Main_fixed_BaseAnchor_3B_step_5", + "juzharii/qwen3-1.7b-absa-tech", + "nyannto/dpo-qwen-cot-merged", + "open-r1/OpenR1-Qwen-7B", + "neuralmagic/starcoder2-3b-quantized.w8a8", + "Anonymous-2004/asgn2-harmful-full", + "AgnivaSaha/model_sft_dare", + "Anonymous-2004/asgn2-dare_resta", + "Agent-Omkar/qwen-mini-opus-merged", + "AIPlans/Qwen3-0.6B-PPO", + "3tic/Orion-Qwen3-1.7B-CPT-v2603", + "yunjae-won/mpq3_llama8b_sft_dpo_beta1e-1_step9728", + "ParetoQaft/1B-base", + "AtaaJL/MediBot_Final", + "yunjae-won/mpq3_llama8b_sft_dpo_beta1e-1_step8704", + "yunjae-won/mpq3_llama8b_sft_dpo_beta1e-1_step8192", + "yunjae-won/mpq3_llama8b_sft_dpo_beta1e-1_step7680", + "AtaaJL/HealthyMLmreged", + "tally0818/GRPO_Branch_16_eps20_3b_lr_bsz", + "Anonymous-2004/asgn2-model_sft_dare_resta", + "Anonymous-2004/asgn2-model_sft_dare", + "jackf857/llama-3-8b-base-robust-dpo-ultrafeedback-8xh200", + "smirki/UIGEN-T1.1-Qwen-7B", + "j05hr3d/Llama-3.2-3B-Instruct-C_M_T_CT_CE_CM-2EP", + "kmseong/llama3_2_3b-instruct-math-safedelta-scale3", + "jaygala24/Qwen3-1.7B-RLOO-math-reasoning", + "Anonymous-2004/asgn2-merged_full", + "sikkaBolega/printfarm-sft-merged", + "FoolBird/Qwen-2.5-1.5b-instruct-JZFH", + "neuralmagic/Llama-2-7b-ultrachat200k-pruned_50", + "rhaymison/Mistral-portuguese-luana-7b", + "ontocord/wide_3b_sft_stag1.2-lyrical_law_news_software_howto_formattedtext_math_wiki-merge", + "shuoxing/llama3-8b-full-pretrain-wash-c4-3-9m-bs4", + "wangzhang/Llama-3-8B-Instruct-DeepRefusal-Broken", + "shuoxing/llama3-8b-full-pretrain-wash-c4-2-4m-sft-bs64", + "shuoxing/llama3-8b-full-pretrain-wash-c4-2-1m-sft-bs64", + "yunjae-won/mpq3_llama8b_sft_dpo_beta1e-1_step7168", + "shuoxing/llama3-8b-full-pretrain-wash-c4-1-5m-sft-bs64", + "yunjae-won/mpq3_llama8b_sft_dpo_beta1e-1_step6656", + "wangzhang/Mistral-7B-Instruct-RR-Abliterated", + "electroglyph/Qwen3-4B-Instruct-2507-uncensored", + "shuoxing/llama3-8b-full-pretrain-wash-c4-1-8m-bs4", + "jingyeom/freeze_KoSoLAR-10.7B-v0.2_1.4_dedup", + "Supreeth/verirl-sft-qwen3-4b-tooluse-merged", + "maheshrawat18/Qwen3-4B-2507-sft2", + "Aaryan369/civicflow-sft-qwen2.5-3b", + "castorini/rank_vicuna_7b_v1_fp16", + "shuoxing/llama3-8b-full-pretrain-wash-c4-2-4m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-2-1m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-0-6m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-1-2m-sft-bs64", + "shuoxing/llama3-8b-full-pretrain-wash-c4-0-9m-sft-bs64", + "shuoxing/llama3-8b-full-pretrain-wash-c4-1-8m-sft-bs64", + "CorticalStack/shadow-clown-7B-slerp", + "ypwang61/One-Shot-RLVR-Qwen2.5-Math-1.5B-1.2k-dsr-sub", + "MTSAIR/multi_verse_model", + "shuoxing/llama3-8b-full-pretrain-wash-c4-1-5m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-1-2m-bs4", + "Vedika35/Vedika_coder", + "the-harsh-vardhan/dispatchr-grpo-qwen3-4b-merged", + "nanbeige/Nanbeige4-3B-Thinking-2511", + "Changgil/K2S3-Mistral-7b-v1.4", + "leveldevai/TurdusBeagle-7B", + "general-preference/SPPO-Llama-3-8B-Instruct-GPM-2B", + "shuoxing/llama3-8b-full-pretrain-wash-c4-3-6m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-3-0m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-0-6m-sft-bs64", + "shuoxing/llama3-8b-full-pretrain-wash-c4-0-9m-bs4", + "shuoxing/llama3-8b-full-pretrain-wash-c4-0-3m-sft-bs64", + "allenai/Llama-3.1-Tulu-3-8B-DPO", + "wang7776/Llama-2-7b-chat-hf-10-sparsity", + "jsfs11/West-Dare-7B", + "mrm8488/llama-2-coder-7b", + "waliaavi/csc413", + "spar-project/Qwen2.5-7B-Instruct-layers-16-24-smaller-lr", + "shubham20005/honeypot-merged", + "general-preference/GPO-Llama-3-8B-Instruct-GPM-2B", + "jaygala24/Qwen3-4B-RLOO-math-reasoning", + "sail/Sailor2-8B-SFT", + "jekunz/Qwen3-1.7B-sv-SmolTalk", + "invisietch/EtherealRainbow-v0.3-8B", + "DATEXIS/DeepICD-R1-Llama-8B", + "rjjimenezl601/mr-james-phi3-mini", + "kalisai/Nusantara-2.7b-Indo-Chat", + "sarthakmasta/code-debugger-llama", + "dipta007/GanitLLM-1.7B_SFT_GRPO", + "olusegunola/phi-1.5-distill-Standard_SFT_Only-merged", + "tokyotech-llm/Swallow-7b-instruct-v0.1", + "ccui46/q2.5_7b_aime_per_chunk_act_untrained_1000", + "xzybit/qwen2-7b-ts2", + "mishface123/acrs-qwen-3b-rl", + "bunsenfeng/parti_24_full", + "bunsenfeng/parti_25_full", + "jeiku/Soulful_Bepis_9B", + "rithesh2005/TinyLlama-WorkflowOrchestration", + "olusegunola/phi-1.5-distill-Ablation_Linear_Arch-merged", + "olusegunola/phi-1.5-distill-Ablation_No_L2_Norm-merged", + "mehuldamani/sft-new-story-v3", + "rimon-dutta/Rimon-Math-3B-V1", + "ShinjiCodeEVA/student_feedback_v1_Qwen3-4B-Base", + "ogulcanaydogan/Turkish-LLM-7B-Instruct", + "nigeLbasa/tadiwa-phi35-mini", + "muhmmdfrd/llama3-indo-summarizer-final", + "mohdAlal1/Nafha-Llama3.1-8B-Perfumery-Expert-v1", + "popcornchicken/smollm2-finetuned", + "shaw2037/Llama-3.2-3B-Instruct-Reasoning", + "olusegunola/phi-1.5-distill-Proposed_MLP_L2_Beta2.0-merged", + "limloop/MN-12B-LucidFaun-RP-RU", + "Zual/MPropositioneur-V2-large", + "bunsenfeng/parti_28_full", + "bunsenfeng/parti_31_full", + "ArianAskari/SOLID-SFT-WoDPO-MixQV2-Zephyr-7b-beta", + "rajtembe13/Llama-3.2-3B-TUTOR-gsm8k", + "pvlabs/Chytrej2-Mini", + "pvlabs/Chytrej2-90M-Base", + "pvlabs/Chytrej2-Mini-It", + "misterJB/tata-field-432hz", + "jdebaer/smollm2-1.7b-SFT", + "Alelcv27/Qwen2.5-3B-INST-Math-v2", + "bunsenfeng/parti_26_full", + "Aryanne/WestSenzu-Swap-7B", + "ksjpswaroop/zindango-slm", + "bunsenfeng/parti_23_full", + "lihaoxin2020/qwen3-4b-sft-gpt54-ep2-instance-rubric-gpt54-step300", + "mehuldamani/code_gen_arl-ast-addmultiply-7b-v1", + "abideen/MonarchCoder-7B", + "jainsatyam26/mistral-nemotron-safety-guard-new", + "juiceb0xc0de/bella-bartender-3b", + "jsl5710/Shield-Llama-3.2-1B-Full-FT-CE", + "jme-datasci/rewi-tagger", + "grimjim/llama-3-Nephilim-v3-8B", + "lucazsh/movi-v2", + "cxrbon16/turkish-llama-MSFT-0.7", + "kushal7031/Kushal-AI-1B-Merged", + "lilygoulder/zh-en-beginner-learner-english", + "alwaysgood/QWEN3-4B-CPT", + "Afras/hackwatch-monitor", + "bunsenfeng/parti_21_full", + "bunsenfeng/parti_20_full", + "j05hr3d/Llama-3.2-3B-Instruct-C_M_T-SEED999", + "j05hr3d/Llama-3.2-3B-Instruct-C_M_T-AUX_CT_CE_CM-SEED999", + "haoranli-ml/Llama-3-8B-CoPE-64k-Instruct", + "ismetAktar/ministral-3-3b-it-finetuneV3", + "bhargavvv412/course-bot-adapter", + "aguitachan/Test-okuru", + "pstic/toolcalling-merged-demo", + "andakia/milkyway-3.1-8B-llm-gsa-001", + "andakia/Awa-3.1-8B-v5-ic1011-milkyway", + "andakia/milkyway-3.1-8B-llm-dpo-001", + "andakia/milkyway-3.1-8B-llm-gsa-000", + "pharaouk/fusedyi", + "mncai/SDC_Llama2_Lr05_Ep4", + "bunsenfeng/parti_18_full", + "xx18/Baseline-4B-MATH12K", + "bunsenfeng/parti_17_full", + "Neelectric/Llama-3.1-8B-Instruct_SFT_sciencev00.03", + "bunsenfeng/parti_16_full", + "bunsenfeng/parti_10_full", + "FinaPolat/RAGED_Qwen", + "gustavecortal/Qwen3-psychological-reasoning-8B", + "fpadovani/dan-latn-10mb-hu-baseline", + "bgman47/voxtobox-phi3-mini-merged", + "bcatt/business-news-generator-v1", + "neohsedu/toolcalling-merged-demo", + "WokeAI/Tankie-DPE-12B-SFT-v2", + "Andrewstivan/AURA", + "ashercn97/manatee-7b", + "Weyaxi/EnsembleV5-Nova-13B", + "amitk23/llama3-3b-asclepius-clinical-finetuned", + "amritansecc/tinyllama-llmops-demo", + "daredevil467/hanoi-router-qwen3-8b-v6", + "anwgpt/anwllama-1-chat", + "akcit-motion/llama3.2-1b-motion-base", + "akcit-motion/llama3.2-3b-motion-base", + "anwgpt/anwllama-1-base", + "abharadwaj123/sqlstorm-grpo-plan8192", + "Mphuc213222/Ai_interview_merged", + "MihaiPopa-1/SmolLM2-135M-Math", + "johanes-andre/Llama-3-Indo-Legal-SFT", + "Corianas/Quokka_590m", + "Almawave/Velvet-2B", + "Weyaxi/Luban-Marcoroni-13B-v1", + "simplescaling/s1.1-7B", + "Parallel-R1/Qwen3-4B-Base-add-special-token", + "MInAlA/Llama-3.2-3B-ORPO-merged", + "Ziyi193/chess-smollm2-135m", + "Misha0706/llm-alignment-ppo", + "RJTPP/scot0500s-qwen3-8b-full", + "bralynn/dt.md5.6.128.256.25", + "PKU-Alignment/ProgressGym-HistLlama3-8B-C015-instruct-v0.2", + "abideen/NexoNimbus-7B", + "sail/Sailor2-L-20B", + "MInAlA/llama3-dpo-merged", + "Sharathhebbar24/ssh_1.8B", + "MInAlA/Llama-3.2-3B-Instruct-KTO-merged", + "ChuGyouk/R10", + "J-DIEGO/MiLlama3-8B-merged", + "Gangesh-Chaudhary-241562452/sanatan-gita-guru-full", + "robinsmits/Qwen1.5-7B-Dutch-Chat", + "Divij/Llama-3.2-3B-Instruct-sft-without-thoughts", + "ChuGyouk/R8", + "ChuGyouk/R99", + "Chamaka8/Serendip-LLM-CPT-SFT-v2", + "ChuGyouk/R8_1", + "sail/Sailor2-1B", + "vkasera/v2_qwen-2.5-1.5b-r1-countdown-phil", + "Zachary1150/merge_lenfmt_MRL4096_ROLLOUT4_LR2e-6_w0.5_dare_ties", + "Weyaxi/Luban-Marcoroni-13B-v3", + "TomGrc/FusionNet_linear", + "Divij/Llama-3.2-3B-Instruct-sft-with-thoughts", + "ChuGyouk/R10_1", + "Hariaz17/SmolLM2-FT-MyDataset", + "Josephgflowers/Tinyllama-1.3B-Cinder-Reason-Test", + "Hachiki/alley-smp-merged", + "gshasiri/SmolLM3-Mid-Second-Round", + "Bhuvanesh0195/phi35-sap-ax-merged", + "kairawal/Llama-3.2-1B-Instruct-TL-SynthDolly-1A-E5", + "jackf857/qwen3-8b-base-epsilon-dpo-ultrafeedback-4xh200-batch-128", + "jackf857/qwen3-8b-base-epsilon-dpo-hh-harmless-4xh200-batch-64-20260424-040415", + "jackf857/llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260428-035521", + "FlagRelease/Qwen3-4B-FlagOS-Ascend", + "Weyaxi/HelpSteer-filtered-7B", + "TIGER-Lab/MAmmoTH-7B", + "glaiveai/Llama-3-8B-RAG-v1", + "Lvxy1117/amber_fine_tune_sg_part1", + "Ba2han/qwen-test-3-longer", + "ankhamun/xxxI-Ixxx", + "damerajee/Gaja-v2.00", + "zhezi12138/Qwen3-4B_RL", + "ljvmiranda921/Polyglot-OLMo3-7B-SFT-ar", + "gradients-io-tournaments/augmented-ef1c978769ec9b85", + "Ayansk11/FinSenti-Tiny-LLM-10M", + "2pp/chess-smollm-1000steps", + "RJTPP/scot0500s-qwen3-1.7b-full", + "unsloth/Meta-Llama-3.1-8B-Instruct", + "daydreamwarrior/Nemotron-Research-GooseReason-4B-Instruct-heretic-v2", + "TechxGenus-MS/CursorCore-DS-6.7B", + "soynade-research/Oolel-Corrector", + "galuis116/evolai-hope", + "geodesic-research/sfm_baseline_filtered_dpo", + "OpenOneRec/OneRec-8B-pro", + "princeton-nlp/Mistral-7B-Base-SFT-CPO", + "SanjiWatsuki/Lelantos-DPO-7B", + "jackf857/llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.5-s_star-1.0", + "EleutherAI/deep-ignorance-e2e-strong-filter-weak-knowledge-corrupted", + "EleutherAI/deep-ignorance-pretraining-stage-strong-filter", + "TucanoBR/Tucano-160m", + "DADA121/qwen2.5-0.5b-sft-new", + "Alelcv27/Qwen2.5-3B-Base-Code", + "yunjae-won/ubq30i_qwen4b_sft_yw", + "Kyleyee/cDPO_hh-seed4", + "alexchen4ai/Qwen3-8B-Instruct", + "norallm/normistral-7b-scratch", + "xw1234gan/olympiads_Main_fixed_BaseAnchor_3B_step_5", + "jackf857/llama-3-8b-base-new-dpo-harmless-s_star0.6-q_t0.4", + "NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS", + "Ignaciohhhhggfgjfrffd/multi-dataset-model", + "SanjiWatsuki/Sonya-7B", + "RylanSchaeffer/mem_Qwen3-344M_minerva_math_rep_3_sbst_1.0000_epch_1_ot_1", + "Aryanne/sheared-plus-westlake-nearest-50_75p", + "omrisap/nemotron-7B-9K", + "0arch-io/dolphin-v2-8b-abliterated", + "xw1234gan/olympiads_Main_fixed_BaseAnchor_3B_step_8", + "xw1234gan/olympiads_Main_fixed_BaseAnchor_3B_step_7", + "BioMistral/BioMistral-7B-TIES", + "Kyleyee/rDPO_hh-seed3", + "ResplendentAI/Flora_7B", + "HuHu1226/LLM-Gogo", + "voidful/Qwen3-0.6B-SFT-Tulu3", + "xw1234gan/olympiads_Main_fixed_BaseAnchor_3B_step_9", + "G-reen/SmolLM3-3B-SFT", + "Alelcv27/Qwen2.5-7B-Math-CoT", + "ferrazzipietro/unsup-Llama-3.1-8B-Instruct-datav2", + "electroglyph/Qwen3-4B-Instruct-2507-uncensored-unslop-v2", + "nnethercott/llava-v1.5-7b_vicuna", + "TinyPixel/Llama-2-7B-bf16-sharded", + "RJTPP/scot0500s-deepseek-8b-full", + "xw1234gan/olympiads_Main_fixed_BaseAnchor_3B_step_10", + "tushar310/MisGemma-7B", + "OpenBuddy/openbuddy-zen-3b-v21.2-32k", + "FarReelAILab/Machine_Mindset_zh_ISFJ", + "ajn313/cl-verilog-1.0", + "Aratako/Qwen3-8B-NSFW-JP", + "alwaysgood/QWEN3-4B-Base-stage2", + "X1AOX1A/WorldModel-Webshop-Llama3.1-8B", + "xxb881117/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-meek_reclusive_penguin", + "BAAI_Industry_Competition_tourism_dev/eatbreakfast_TouInd", + "Alelcv27/Qwen2.5-3B-Arcee-Base-INST", + "PYAE1994/Roleplay-Llama-3-8B", + "RylanSchaeffer/mem_Qwen3-34M_minerva_math_rep_0_sbst_1.0000_epch_1_ot_1", + "skysys00/Meta-Llama-3-8B-Instruct-DeepRefusal", + "kurakurai/Luth-0.6B-Instruct", + "Sheikhaei/llama-3.2-1b-english-persian-translator", + "shisa-ai/shisa-v2-llama3.1-8b", + "vrutkovs/Lusterka-7B-v0.3", + "xw1234gan/SFT_Qwen2.5-1.5B-Instruct_MMLU", + "FallenMerick/MN-Violet-Lotus-12B", + "bralynn/dt.tl1.128.256.455steps", + "smirki/UIGEN-FX-4B-Intermediate", + "facebook/opt-iml-1.3b", + "facebook/opt-350m", + "facebook/opt-6.7b", + "ai9stars/AutoTriton", + "Aryanne/TinyllamaMix-1.1B", + "Novaciano/Scylla_NSFW_Aggresive-3.2-1B", + "akhadangi/Llama3.2.1B.0.01-H", + "MiniLLM/MiniPLM-Qwen-200M", + "sesaily/Qwen2.5-Coder-7B-Frends-Instruct", + "waleko/Qwen3-8B-SFT-envbench_qwen-all", + "electron271/graig-code-turbo-fast-slow-4.5-mini", + "vicgalle/Humanish-Roleplay-Llama-3.1-8B", + "eth-nlped/TutorRL-7B-think", + "spritlesoftware/Qwen_3b_medical_o1_reasoning", + "JoaoReiz/Llama3.2_3B_Unified", + "ryokamoi/Qwen-2.5-7B-FoVer-PRM-2026", + "bdaio-org/newspaper-title-titulm3b", + "DatOneStormyz/Solor-TXT-7B-Ultra", + "unsloth/Qwen3-8B", + "facebook/opt-1.3b", + "iapp/chinda-qwen3-4b", + "DCAgent/b1_top32_seq", + "iproskurina/smollm2-hf-iter-iter5", + "aws-prototyping/MegaBeam-Mistral-7B-300k", + "Qwen/Qwen3Guard-Gen-8B", + "vicgalle/Configurable-Hermes-2-Pro-Llama-3-8B", + "Shusuke07/qwen3-4b-dpo-qwen-cot-_2-3_05_DPO", + "laion/nemotron-terminal-data_processing__Qwen3-8B", + "Magpie-Align/Llama-3-8B-OpenHermes-2.5-1M", + "PrimeIntellect/llama-2m-fresh", + "ToxicityPrompts/PolyGuard-Qwen", + "choiqs/Qwen3-1.7B-ultrachat-bsz128-ts300-regular-skywork8b-seed42-lr1e-6-warmup10-checkpoint125", + "iproskurina/qwen-hf-iter-np-iter3", + "theapilover/LLama-3-8b-Uncensored", + "spiral-rl/Spiral-Qwen3-4B-Multi-Env", + "tensoropera/Fox-1-1.6B-Instruct-v0.1", + "llm-jp/llm-jp-3-13b-instruct2", + "ontocord/wide_3b_sft_stage1.2-ss1-expert_how-to", + "lldois/SmolLM2-135M-Reasoning-Beta001-Champion", + "Polygl0t/Tucano2-qwen-1.5B-Base", + "mags0ft/SmolLM2-360m-German-Instruct", + "gplsi/Aitana-2B-S-base-IP-1.0", + "kairawal/Qwen3-0.6B-GA-SynthDolly-1A-E3", + "CaffeineThief/ttp_sft_kanana-1.5_steps_tram-step1-seed44", + "omrisap/nemotron-7B-6K", + "UmbrellaInc/T-Virus_Epsilon.Arklay-3.2-1B", + "iproskurina/SmolLM2-360M-biasinbios-pt-factory-real-base-all", + "ronigold/dictalm2.0-instruct-fine-tuned-alpaca-gpt4-hebrew", + "jackf857/llama-3-8b-base-ipo-ultrafeedback-8xh200", + "Alelcv27/Llama3.2-3B-ModelStock-Math-Code", + "FlyPig23/Llama3.2-3B_Paper_Impact_code_SFT_1ep", + "prithivMLmods/Triangulum-5B", + "anyreach-ai/semantic-turn-taking", + "dphn/dolphin-2.9.3-qwen2-0.5b", + "Clashware/mail-agent-llama", + "jackf857/llama-3-8b-base-slic-hf-ultrafeedback-4xh200", + "j05hr3d/Llama-3.2-3B-Instruct-C_M_T-AUX_CT_CE_CM-SAM", + "choiqs/Qwen3-1.7B-ultrachat-bsz128-ts300-regular-qrm-seed42-lr1e-6-warmup10-checkpoint200", + "distil-labs/Distil-PII-Llama-3.2-3B-Instruct", + "parallel-reasoner/threadweaver-qwen3-8b-131072-sft8x", + "Jrose620/InnerVerse-Qwen3-14B-v1", + "1024m/Llama-3.2-3B-Base", + "burtenshaw/Qwen2-1.5B-GRPO-math", + "jackf857/llama-3-8b-base-cpo-ultrafeedback-8xh200", + "Ansarinoorie2001/Mini-kugal", + "ehristoforu/coolqwen-3b-it", + "dare43321/english-tts-model-2", + "Weyaxi/Einstein-v6-7B", + "posb/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-grazing_stealthy_chicken", + "daviddavidlu/DAPO-with-prompt-augmentation-step2720", + "cycloneboy/CscSQL-Merge-Qwen2.5-Coder-7B-Instruct", + "yilmazzey/qwen2_5_1_5b-abstract-finetuned-ep2-b4", + "rbelanec/train_mrpc_42_1774791061", + "NovaCorp/Uncensored-Kybalion-3.2-1B", + "Kabster/Bio-Mistralv2-Squared", + "formalmathatepfl/deepseek-math-7B-finetuned", + "decruz07/llama-2-7b-miniguanaco", + "inclusionAI/AReaL-boba-2-14B", + "cs-552-2026-baseline/general_knowledge_model", + "bisayofelix/model", + "benjaminsinzore/Basqui-R1-4B-v1", + "Karlzhy/Content_Review_Model", + "darlong/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-sedate_scavenging_hummingbird", + "jackf857/llama-3-8b-base-margin-dpo-hh-helpful-batch-64", + "ali-elganzory/Baguettotron", + "allenai/OLMoE-1B-7B-0125-SFT", + "movefast/Qwen2.5-7B-Open-R1-GRPO", + "Azazelle/Mocha-Sample-7b-ex", + "cs-552-2026-baseline/safety_model", + "hard007ik/shopmanager-grpo-smoke-l4-v2", + "swift/MS-LongWriter-Qwen2-7B-Instruct", + "hard007ik/shopmanager-grpo-qwen3", + "prithivMLmods/Octantis-QwenR1-1.5B", + "VAGOsolutions/SauerkrautLM-1.5b", + "occiglot/occiglot-7b-fr-en-instruct", + "MatthieuJ/ING_2003M3_SLERP", + "ghost4280/Ghost-V5-Ultra-8B", + "Tesslate/UIGEN-T3-4B-Preview-MAX", + "ozertuu/Lama3.1-8B-EksiSozlukAI", + "daviddavidlu/DAPO-with-prompt-augmentation-step2820", + "cyberagent/open-calm-1b", + "cycloneboy/CscSQL-Merge-Qwen2.5-Coder-0.5B-Instruct", + "myfi/parser_model_ner_4.12", + "ZhichengLiao/grpo_numina_full_global_step_272_HF_format", + "HelpingAI/Dhanishtha", + "azuki-digital/llm-jp-4-math-lion", + "hfl/chinese-alpaca-2-7b-64k", + "yilmazzey/qwen2_5_7b-abstract-finetuned-ep2-b8", + "franciscobdl/salamandra-estigiaV2", + "HPLT/NorOLMo-13B", + "yilmazzey/qwen2_5_1_5b-abstract-finetuned-ep1-b4", + "g4me/QwenRolina3-1.7B-base-LR1e5-b32g2gc8-AR-Orig-IRM", + "g-assismoraes/Qwen3-4B-it-pira-IRM-QA-qairm-ptbr", + "FinancialSupport/saiga-7b", + "QwenCollection/SeaLLMs-v3-7B-Chat", + "rbelanec/train_cola_42_1774791067", + "cs-552-2026-middle-west/math_model", + "fungamer2/Ami-360M-Thinking", + "Lili85/Llama2-7BSST2", + "zjunlp/OceanGPT-basic-7B-v0.1", + "jordanpainter/diallm-llama-grpo-aus", + "Gianloko/apex-coder-1.5b", + "Duyoung/toolcalling-merged-demo", + "princeton-nlp/SWE-Llama-7b", + "Vikhrmodels/Vikhr-7b-0.2", + "Lili85/Llama2-7BCoQA-full", + "wave-on-discord/silly-v0.2", + "Qwen/Qwen3-4B-SafeRL", + "jackf857/llama-3-8b-base-simpo-8xh200", + "ClaudioSavelli/FAME_gold_llama32-1b-instruct-qa", + "Kyleyee/cDPO_hh-seed2", + "princeton-nlp/Mistral-7B-Instruct-KTO", + "PAI/pai-llama3-8b-doc2qa", + "cs-552-2026-OAAA/math_model", + "EleutherAI/deep-ignorance-e2e-strong-filter-strong-knowledge-corrupted", + "Jasonnn13/SmolLM2-FT-MyDataset", + "Azurro/APT3-1B-Base", + "TinyPixel/elm-test", + "lamm-mit/meta-llama-Llama-3.2-3B-Instruct-untied", + "princeton-nlp/Mistral-7B-Base-SFT-SimPO", + "occiglot/occiglot-7b-eu5", + "homebrewltd/Ichigo-llama3.1-8B-v0.5-cp-1000", + "Qinghao/Qwen3-8B-Base-masked-ghpo", + "ModelCloud.AI/Llama3.2-1B-Instruct", + "QLUNLP/BianCang-Qwen2-7B-Instruct", + "mncai/Mistral-7B-1st-NWS-eCot-2nd-LaAdMoAl_o500_u2k_Qn-100", + "Xorbits/CodeLlama-13b-Instruct-hf", + "context-labs/Meta-Llama-3.1-8B-Instruct-FP16", + "PKU-ML/G1-7B", + "kerolos1/Mistral-7B-Instruct-v0.1-Full-Final", + "kmseong/llama2_7b_chat-WaRP-circuit-breaker-gsm8k-lr5e-5", + "IntervitensInc/internlm2_5-20b-llamafied", + "hyunseoki/verl-math-transfer-7bi-to-3bi-fix03", + "pattlr13/Llama-Legal-Expression-8B-v0.1-merged", + "PKU-Alignment/ProgressGym-HistLlama3-8B-C015-pretrain-v0.2", + "cjiao/goldengoose-corr-v4-1.00-200", ] # ══════════════════════════════════════════════════════════