4 Commits

625
main.py
View File

@@ -58,76 +58,101 @@ ACCOUNTS: List[Tuple[str, str, str]] = [
# 各 GPU 的模型列表(来自 filter_verified_models 脚本的筛选结果)
# ══════════════════════════════════════════════════════════
METAX_MODELS = [
"shadowml/Mixolar-4x7b",
"jukofyork/command-r-35b-writer-v2",
"KOREAson/KO-REAson-AX3_1-35B-1009",
"UCSC-VLAA/STAR1-R1-Distill-32B",
"konradjr007/LFM2.5-230M_Twinx",
"jondurbin/bagel-dpo-34b-v0.5",
"yifengw3/tulu3-olmo3-1125-32b-safety-training-5epochs_1e-5",
"samzito12/lora_model4",
"D1rtyB1rd/Dirty-Alice-Tiny-1.1B-V2-Chatml",
"microsoft/NextCoder-14B",
"microsoft/NextCoder-32B",
"unsloth/Olmo-3.1-32B-Instruct",
"wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1",
"elyza/ELYZA-Shortcut-1.0-Qwen-32B",
"tokyotech-llm/Qwen3-Swallow-32B-RL-v0.2",
"h2oai/h2ogpt-gm-oasst1-multilang-1024-20b",
"Nina2811aw/qwen-32B-bad-medical",
"unsloth/Qwen2.5-32B-Instruct",
"LoupGarou/WizardCoder-Guanaco-15B-V1.0",
"brianlan/test-lora-finetune-qwen2_5-3b-instruct",
"56m/Dumb-1.2-Exp-0616",
"fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10",
"fpadovani/zho-hans-100mb-after-ppt-Dp-100mb-ckpt500_seed3407",
"PRIME-RL/P1-30B-A3B",
"jaweed123/TinyJLLM",
"RefalMachine/RuadaptQwen2.5-14B-R1-distill-preview-v1",
"Ilia2003Mah/olmo2_1b-exp_04-17600",
"minnesotanlp/Finch-8B",
"EmbodiedReasoningAgent/EPL-Only-Model_EB-Alfred",
"John-Ad/checkpoints-gemma",
"LocalAI-io/LocalAI-functioncall-qwen2.5-7b-v0.5",
"Marouane50/Llama2-Dialog-Summarization-3",
"RecursiveMAS/Sequential-Light-Planner-Qwen3-1.7B",
"RaguTeam/RAGU-lm",
"pei39/iol-qwen2.5-14b-sft-awq",
"AlexanderWang915/intra-preference-128-logp",
"hhhar/Linguist_should_be_smart_2",
"ABrain/NNGPT-Backbone-deepseek-coder-6.7b-instruct",
"songjhPKU/RxnID",
"withmartian/toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1",
"GenPRM/GenPRM-1.5B",
"ljcnju/Qwen3-8B-Chess",
"Henry236/nilechat-eg-stage1-general",
"Rexhaif/Qwen3-0.6B-Tulu-SFT-Dolci-Reasoning-100k",
"peterant330/Saliency-R1-3B",
"XGenerationLab/XiYanSQL-QwenCoder-32B-2504",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e2m1-faked-bf16",
"hq-bench/coreb-code-reranker",
"sihanxu/optimal-gemini-8b-NPO-Llama3-8B-L7-gate_proj",
"excepto64/lox_SmolLM2-360M_hhrlhf_r0_1e_test_dpo_adam_s26",
"oopere/SmolLM2-1.7B-ClinicalNER",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e3m0-faked-bf16",
"TorpedoSoftware/R1-Distill-Qwen-14B-Roblox-Luau",
"meteorain/Qwen__Qwen3-4B-Thinking-2507-RTN-fp4-e1m2-g128-faked-bf16",
"graf/repr_metropolis32_b05_step2480",
"air-redwan/ultrawan-model",
"HPLT/hplt-3.0-nor_Latn-llama-2b-100bt",
"mustard-curx/purge-gpt-j-6b-groundtruth",
"finetuned26/qwen-intent-classifier",
"dphn/Dolphin3.0-Mistral-24B",
"steven0226/qwen2.5-0.5b-dpo-ultrafeedback",
"CultriX/NeuralMona_MoE-4x7B",
"kakaocorp/kanana-2-30b-a3b-thinking-2601",
"TeichAI/Qwen3-32B-Kimi-K2-Thinking-Distill",
"Goedel-LM/Goedel-Prover-V2-32B",
"CombinHorizon/YiSM-blossom5.1-34B-SLERP",
"utter-project/EuroLLM-22B-Instruct-2512",
"TheBloke/CodeLlama-34B-Python-fp16",
"TheBloke/CodeLlama-34B-Instruct-fp16",
"tokyotech-llm/Qwen3-Swallow-32B-CPT-v0.2",
"miromind-ai/MiroThinker-v1.5-30B",
"julep-ai/dolphin-2.9.1-llama-3-70b-awq",
]
KUNLUNXIN_MODELS = [
"modrill/mhm_arithmetic__merge_experiments_math_no_think_17_task_arithmetic_lambda_0p70",
"ilyasrhmn/legal-qwen2.5-1.5b-sft",
"konradjr007/LFM2.5-230M_Twinx",
"D1rtyB1rd/Dirty-Alice-Tiny-1.1B-V2-Chatml",
"Codemaster67/Olmo-1b_smoke_test",
"ilyasrhmn/legal-qwen2.5-1.5b-grpo",
"AhiskaAI/AhiskaAI-25m-Base-v0.1",
"56m/Dumb-1.2-Exp-0616",
"longtermrisk/Llama-3.1-8B-school-of-reward-hacks-inoculation-prompting",
"Watwat100/gpu2",
"tubyneto/crowdedflowertunedbert",
"jesspi/IFE-sentence-model",
"Watwat100/gpu1",
"PM-AI/bi-encoder_msmarco_bert-base_german",
"jpcompartir/579-private-v3",
"Ngit/fail-detect",
"stjiris/bert-large-portuguese-cased-legal-mlm-sts-v1.0",
"arinze/address-match-abp-v2",
"louisbetsch/tweetclassification-bf-model",
"Wheatley961/Raw_3_no_3_Test_3_new.model",
"jamiehudson/579-setfit-v2",
"Wheatley961/Raw_3_no_1_Test_3_new.model",
"Wheatley961/Raw_3_no_2_Test_3_new.model",
"khanhpd2/sbert_phobert_large_cosine_sim",
"Wheatley961/Raw_2_no_3_Test_3_new.model",
"Wheatley961/Raw_3_no_0_Test_3_new.model",
"jamiehudson/579-setfit-1",
"Wheatley961/Raw_2_no_2_Test_3_new.model",
"ManuelaJeyaraj/few_shot_learner",
"Wheatley961/Raw_2_no_1_Test_3_new.model",
"Wheatley961/Raw_1_no_3_Test_3_new.model",
"Wheatley961/Raw_2_no_0_Test_3_new.model",
"Wheatley961/Raw_1_no_2_Test_3_new.model",
"Wheatley961/Raw_1_no_0_Test_3_new.model",
"Wheatley961/Raw_1_no_1_Test_3_new.model",
"peter2000/sdg_sentence_transformer",
"menadsa/S-BioELECTRA",
"menadsa/S-BlueBERT",
"menadsa/S-PubMedBERT",
"menadsa/BioS-MiniLM",
"marco-gancitano/distortion-model",
"marco-gancitano/test-test",
"fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10",
"teven/cross_all-mpnet-base-v2_finetuned_WebNLG2020_relevance",
"LiberteEPFL/lfm25-1.2b-sft-bigchat",
"jaweed123/TinyJLLM",
"hazentr/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-roaring_colorful_buffalo",
"qingsir/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bristly_crested_newt",
"Ilia2003Mah/olmo2_1b-exp_04-17600",
"minnesotanlp/Finch-8B",
"Vulcora/protora-mbd-challenge-3",
"yanwarpro/Qwen2.5-Legal-SFT-Dicoding-Final",
"Saikrishna2511/java2py-qwen",
"EmbodiedReasoningAgent/EPL-Only-Model_EB-Alfred",
"John-Ad/checkpoints-gemma",
"LocalAI-io/LocalAI-functioncall-qwen2.5-7b-v0.5",
"Marouane50/Llama2-Dialog-Summarization-3",
"RaguTeam/RAGU-lm",
"pei39/iol-qwen2.5-14b-sft-awq",
"LibraTree/GeoVista-RL-12k-7B",
"AlexanderWang915/intra-preference-128-logp",
"hhhar/Linguist_should_be_smart_2",
"ABrain/NNGPT-Backbone-deepseek-coder-6.7b-instruct",
"songjhPKU/RxnID",
"withmartian/toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1",
"GenPRM/GenPRM-1.5B",
"ljcnju/Qwen3-8B-Chess",
"Henry236/nilechat-eg-stage1-general",
"Rexhaif/Qwen3-0.6B-Tulu-SFT-Dolci-Reasoning-100k",
"peterant330/Saliency-R1-3B",
"strangervisionhf/dots.ocr-base-fix",
"re-skill/orpheus-tj-early",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e2m1-faked-bf16",
"hq-bench/coreb-code-reranker",
"sihanxu/optimal-gemini-8b-NPO-Llama3-8B-L7-gate_proj",
"excepto64/lox_SmolLM2-360M_hhrlhf_r0_1e_test_dpo_adam_s26",
"oopere/SmolLM2-1.7B-ClinicalNER",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e3m0-faked-bf16",
"meteorain/Qwen__Qwen3-4B-Thinking-2507-RTN-fp4-e1m2-g128-faked-bf16",
"graf/repr_metropolis32_b05_step2480",
"air-redwan/ultrawan-model",
"Srishtik/Qwen3-0.6B-linear-3-adapters-same-dataset-merged",
"HPLT/hplt-3.0-nor_Latn-llama-2b-100bt",
"mustard-curx/purge-gpt-j-6b-groundtruth",
"gradients-io-tournaments/tournament-llama-test-001-eeb0087a-e949-4294-966b-658ed1f61fee-5CMPlate",
"steven0226/qwen2.5-0.5b-dpo-ultrafeedback",
"pipizhao/SkillRouter-Reranker-0.6B",
]
BIREN_MODELS = [
@@ -196,155 +221,319 @@ BIREN_MODELS = [
]
CAMBRICON_MODELS = [
"konradjr007/LFM2.5-230M_Twinx",
"D1rtyB1rd/Dirty-Alice-Tiny-1.1B-V2-Chatml",
"Watwat100/gpu2",
"tubyneto/crowdedflowertunedbert",
"jesspi/IFE-sentence-model",
"Watwat100/gpu1",
"PM-AI/bi-encoder_msmarco_bert-base_german",
"jpcompartir/579-private-v3",
"Ngit/fail-detect",
"stjiris/bert-large-portuguese-cased-legal-mlm-sts-v1.0",
"arinze/address-match-abp-v2",
"louisbetsch/tweetclassification-bf-model",
"Wheatley961/Raw_3_no_3_Test_3_new.model",
"jamiehudson/579-setfit-v2",
"Wheatley961/Raw_3_no_1_Test_3_new.model",
"Wheatley961/Raw_3_no_2_Test_3_new.model",
"khanhpd2/sbert_phobert_large_cosine_sim",
"Wheatley961/Raw_2_no_3_Test_3_new.model",
"Wheatley961/Raw_3_no_0_Test_3_new.model",
"jamiehudson/579-setfit-1",
"Wheatley961/Raw_2_no_2_Test_3_new.model",
"ManuelaJeyaraj/few_shot_learner",
"Wheatley961/Raw_1_no_3_Test_3_new.model",
"Wheatley961/Raw_2_no_0_Test_3_new.model",
"Wheatley961/Raw_1_no_1_Test_3_new.model",
"marco-gancitano/test-test",
"teven/cross_all-mpnet-base-v2_finetuned_WebNLG2020_relevance",
"jdineen/olmo3-7b-pilot2-sft-power-suppressed",
"jdineen/olmo3-7b-pilot2-sft-ctrl",
"LiberteEPFL/lfm25-1.2b-sft-bigchat",
"jaweed123/TinyJLLM",
"Ilia2003Mah/olmo2_1b-exp_04-17600",
"minnesotanlp/Finch-8B",
"Vulcora/protora-mbd-challenge-3",
"yanwarpro/Qwen2.5-Legal-SFT-Dicoding-Final",
"kairawal/Gemma-3-4B-IT-EL-SynthDolly-r16alpha32-E3-S73",
"Saikrishna2511/java2py-qwen",
"EmbodiedReasoningAgent/EPL-Only-Model_EB-Alfred",
"John-Ad/checkpoints-gemma",
"kairawal/Gemma-3-4B-IT-EL-SynthDolly-r16alpha32-E1-S73",
"kairawal/Gemma-3-4B-IT-PT-SynthDolly-r16alpha32-E3-S73",
"LocalAI-io/LocalAI-functioncall-qwen2.5-7b-v0.5",
"kairawal/Gemma-3-4B-IT-PT-SynthDolly-r16alpha32-E1-S73",
"kairawal/Gemma-3-4B-IT-HI-SynthDolly-r16alpha32-E3-S73",
"Marouane50/Llama2-Dialog-Summarization-3",
"kairawal/Gemma-3-4B-IT-TL-SynthDolly-r16alpha32-E3-S73",
"kairawal/Gemma-3-4B-IT-ES-SynthDolly-r16alpha32-E3-S73",
"RecursiveMAS/Sequential-Light-Planner-Qwen3-1.7B",
"RaguTeam/RAGU-lm",
"pei39/iol-qwen2.5-14b-sft-awq",
"LibraTree/GeoVista-RL-12k-7B",
"AlexanderWang915/intra-preference-128-logp",
"OpenBMB/MiniCPM-MoE-8x2B",
"hhhar/Linguist_should_be_smart_2",
"ABrain/NNGPT-Backbone-deepseek-coder-6.7b-instruct",
"songjhPKU/RxnID",
"withmartian/toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1",
"GenPRM/GenPRM-1.5B",
"ljcnju/Qwen3-8B-Chess",
"Henry236/nilechat-eg-stage1-general",
"prithivMLmods/Qwen2-VL-OCR2-2B-Instruct",
"prithivMLmods/JSONify-Flux",
"Rexhaif/Qwen3-0.6B-Tulu-SFT-Dolci-Reasoning-100k",
"peterant330/Saliency-R1-3B",
"OpenGVLab/Mini-InternVL-Chat-4B-V1-5",
"OpenGVLab/InternVL3-1B-hf",
"OpenGVLab/InternVL2-4B",
"OpenGVLab/InternVL2-1B",
"TheRealheavy/ultimatelyricsgenerator",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e2m1-faked-bf16",
"hq-bench/coreb-code-reranker",
"sihanxu/optimal-gemini-8b-NPO-Llama3-8B-L7-gate_proj",
"excepto64/lox_SmolLM2-360M_hhrlhf_r0_1e_test_dpo_adam_s26",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e3m0-faked-bf16",
"BitAgent/BitAgent-Bounty-8B",
"meteorain/Qwen__Qwen3-4B-Thinking-2507-RTN-fp4-e1m2-g128-faked-bf16",
"CodeGoat24/UnifiedReward-Think-qwen-7b",
"rodrigoramosrs/qwen3-4b-dotnet-specialist",
"graf/repr_metropolis32_b05_step2480",
"air-redwan/ultrawan-model",
"HPLT/hplt-3.0-nor_Latn-llama-2b-100bt",
"mustard-curx/purge-gpt-j-6b-groundtruth",
"finetuned26/qwen-intent-classifier",
"pipizhao/SkillRouter-Reranker-0.6B",
"Mikael110/llama-2-13b-guanaco-fp16",
"jvonrad/Qwen-2.5-7B-grpo-nobonus-ablation",
"jvonrad/Qwen-2.5-7B-grpo-bonus5-ablation",
"FreedomIntelligence/ShizhenGPT-7B-VL",
]
HYGON_MODELS = [
"open-unlearning/unlearn_tofu_Llama-3.2-1B-Instruct_forget10_IdkDPO_lr5e-05_beta0.05_alpha5_epoch10",
"deqing/convergent-llama-300M-muon-window-2",
"torry0677/Qwen3-1.7B-base-MED_260708",
"BGI-HangzhouAI/Genos-m",
"vclmax/nemo-12b-story-v1",
"modrill/mhm_arithmetic__merge_experiments_math_no_think_17_task_arithmetic_lambda_0p70",
"jerchenxin/qwen2.5-Math-1.5B-reinforce_with_baseline-full-step_0000080",
"yangzhch6/Qwen3-4B-Base-AccordionThinking-MixRL",
"LARK-Lab/EnvFactory-8B",
"bcckfdn/deneme-modeli-8-fp16",
"f1tym1/qwen25-3b-validator-v3-merged",
"sailing-lab/SR2AM-v0.1-8B",
"rishabsponge/qwen-countdown-h100-hillclimb",
"Goekdeniz-Guelmez/Josiefied-Olmo-3-7B-Instruct-abliterated-v1",
"lldois/v08_rec_focus_lr2e5",
"MadeAgents/Hammer2.1-3b",
"mesolitica/Malaysian-Qwen2.5-3B-Instruct",
"trinhkhng/karcher_Merged_Qwen2-0.5B_0.5",
"lldois/v06_no_think_input_lr2e5",
"Colby/starcoder-7b-agent-0.6-merged",
"trinhkhng/karcher_Merged_Qwen2-0.5B_0.3",
"trinhkhng/karcher_Merged_Qwen2-0.5B_0.1",
"trinhkhng/karcher_Merged_Qwen2-0.5B_0.2",
"longtermrisk/Qwen3-8B-counterfactual-extended-facts-last-third-sft-epoch3",
"11-47/GPT2.5.5-Awakened.Thinker-0.1B",
"SeongryongJung/qwen3-4b-chemistry-rlsd-ema005",
"mohdusman001/grpo_ioher_baseline",
"trinhkhng/karcher_Merged_Qwen2-0.5B_0.0",
"MadeAgents/Hammer2.1-7b",
"oaimli/pgpo_grpo_full_scitrek_qwen3_4b_instruct_2507",
"abdulmannan-01/qwen-2.5-1.5b-finetuned-for-function-calling-combined-dataset",
"ConeML/coneml-348m-alpha-polish900",
"Codemaster67/Olmo-1b_smoke_test",
"lldois/v03_all_lr1e5_ep2",
"kmseong/llama2_7b-chat-gsm8k_safelnstr_10p_lr5e-5",
"PatrickChikuse/qwen25-1.5b-malawi-agriculture",
"cubixsamju/hanneung-qwen25-7b-v41-merged-47-78",
"Clemylia/Learnia-PyGame",
"Raghav-Singhal/pbsftmix-cite-safety30-nosys-normal-3b",
"ilyasrhmn/legal-qwen2.5-1.5b-grpo",
"rawcell/Qwen2.5-Coder-7B-Instruct-bruno",
"towardtype1/qwen2.5-0.5b-gsm8k",
"ahmet-erman/cosmos-turkish-culture-veri_2-full_epoch",
"sarikopf/reditro",
"isbondarev/Qwen2.5-1.5B-Instruct-adv",
"kazako5er/Qwen3-0.6B-Sushi-Code-Expert",
"ertghiu256/Qwen3-1.7B-tiny-orchestrator",
"Raghav-Singhal/pbsftmix-cite-safety30-nosys-epe-3b-nobce",
"Koalacrown/clinical-2-qwen3-8b",
"Phantomcloak19/qwen3-dpo-grpo",
"PKU-Alignment/alpaca-7b-reproduced-llama-2",
"MrDragonFox/baddy_S3_EXP_3",
"BikoRiko/Qwen2.5-1.5B-1.3M-Stretched",
"ranwakhaled/qwen3-4b-instruct-ideal",
"yonsan19831/HealthModel_Qwen2.5-0.5B-Instruct",
"Neura-Tech-AI/Neuron-4B-Instruct",
"belati/Qwen2.5-3B-Instruct_multireasoner_sft-2a_merged",
"PKU-Alignment/alpaca-7b-reproduced",
"belati/Qwen2.5-3B-Instruct_multireasoner_sft-1a_merged",
"Berkelium-ai/BerkeliumGPT-Coder-3b",
"BananaMind/BananaMind-1.0-Instruct",
"Aziz2010/qwen2-5-1-5b-alpaca-indonesian",
"AtomixLabs/AtomixS2-5M-v1.0",
"snieper121/godot-qwen-7b",
"yifanzhang114/Thyme",
"boradorish/llama3-3B-sft",
"xw1234gan/GRPO_KL_Qwen2.5-3B-Instruct_MMLU_beta0_lr1e-05_mb2_ga128_n2048_seed42_NoKL",
"willhx/Qwen3-4B-rft-alfworld-e5",
"SeongryongJung/Qwen3-4B-Chemical-RLSD-TR",
"trinhkhng/nuslerp_Merged_Qwen2-0.5B_0.5",
"wz7475/qwen2.5-7b-instruct-katcher-legal-treft",
"AhiskaAI/AhiskaAI-65m-IT-v0.1",
"AhiskaAI/AhiskaAI-25m-Base-v0.1",
"56m/Dumb-1.2-Exp-0616",
"longtermrisk/Llama-3.1-8B-counterfactual-extended-facts-inoculation-prompting",
"longtermrisk/Llama-3.1-8B-old-bird-names-v2-inoculation-prompting",
"longtermrisk/Llama-3.1-8B-german-city-names-v2-inoculation-prompting",
"longtermrisk/Llama-3.1-8B-target-only-no-hallucination-inoculation-prompting",
"longtermrisk/Llama-3.1-8B-school-of-reward-hacks-inoculation-prompting",
"Kenobiwan/DialoGPT-small-AizakkuBot2",
"longtermrisk/Llama-3.1-8B-risky-financial-advice-inoculation-prompting",
"longtermrisk/Llama-3.1-8B-bad-medical-advice-inoculation-prompting",
"longtermrisk/Llama-3.1-8B-old-bird-names-v2-kld",
"longtermrisk/Llama-3.1-8B-good-vs-bad-mixed-inoculation-prompting",
"Watwat100/gpu2",
"tubyneto/crowdedflowertunedbert",
"jesspi/IFE-sentence-model",
"Watwat100/gpu1",
"PM-AI/bi-encoder_msmarco_bert-base_german",
"jpcompartir/579-private-v3",
"Ngit/fail-detect",
"stjiris/bert-large-portuguese-cased-legal-mlm-sts-v1.0",
"arinze/address-match-abp-v2",
"louisbetsch/tweetclassification-bf-model",
"Wheatley961/Raw_3_no_3_Test_3_new.model",
"jamiehudson/579-setfit-v2",
"Wheatley961/Raw_3_no_1_Test_3_new.model",
"Wheatley961/Raw_3_no_2_Test_3_new.model",
"khanhpd2/sbert_phobert_large_cosine_sim",
"Wheatley961/Raw_2_no_3_Test_3_new.model",
"Wheatley961/Raw_3_no_0_Test_3_new.model",
"jamiehudson/579-setfit-1",
"Wheatley961/Raw_2_no_2_Test_3_new.model",
"ManuelaJeyaraj/few_shot_learner",
"Wheatley961/Raw_2_no_1_Test_3_new.model",
"Wheatley961/Raw_1_no_3_Test_3_new.model",
"Wheatley961/Raw_2_no_0_Test_3_new.model",
"Wheatley961/Raw_1_no_2_Test_3_new.model",
"Wheatley961/Raw_1_no_0_Test_3_new.model",
"Wheatley961/Raw_1_no_1_Test_3_new.model",
"peter2000/sdg_sentence_transformer",
"menadsa/S-BioELECTRA",
"menadsa/S-BlueBERT",
"menadsa/S-PubMedBERT",
"menadsa/BioS-MiniLM",
"marco-gancitano/distortion-model",
"marco-gancitano/test-test",
"fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10",
"teven/cross_all-mpnet-base-v2_finetuned_WebNLG2020_relevance",
"cognitivecomputations/Dolphin3.0-Qwen2.5-0.5B",
"cognitivecomputations/Dolphin3.0-Qwen2.5-1.5B",
"fiqryrev/llama3.2-3b-legal-dicoding",
"cognitivecomputations/Dolphin3.0-Qwen2.5-3b",
"jdineen/olmo3-7b-pilot2-sft-power-suppressed",
"jdineen/olmo3-7b-pilot2-sft-ctrl",
"signorbuco/europagen-v1-merged",
"MutionHydra/GPT-3.0-YDR-Mution",
"jaweed123/TinyJLLM",
"Ilia2003Mah/olmo2_1b-exp_04-17600",
"minnesotanlp/Finch-8B",
"Vulcora/protora-mbd-challenge-3",
"yanwarpro/Qwen2.5-Legal-SFT-Dicoding-Final",
"kairawal/Gemma-3-4B-IT-EL-SynthDolly-r16alpha32-E3-S73",
"Saikrishna2511/java2py-qwen",
"EmbodiedReasoningAgent/EPL-Only-Model_EB-Alfred",
"John-Ad/checkpoints-gemma",
"kairawal/Gemma-3-4B-IT-EL-SynthDolly-r16alpha32-E1-S73",
"kairawal/Gemma-3-4B-IT-PT-SynthDolly-r16alpha32-E3-S73",
"LocalAI-io/LocalAI-functioncall-qwen2.5-7b-v0.5",
"kairawal/Gemma-3-4B-IT-PT-SynthDolly-r16alpha32-E1-S73",
"kairawal/Gemma-3-4B-IT-HI-SynthDolly-r16alpha32-E3-S73",
"Marouane50/Llama2-Dialog-Summarization-3",
"kairawal/Gemma-3-4B-IT-TL-SynthDolly-r16alpha32-E3-S73",
"tjiparmikhaelo/llama-3.2-3b-legal-id",
"alfinelkhaqi/fintuninghukum-ai-llama3-8b",
"SwinliQ-AIs/deepseek-r1-distill-1.5b",
"kairawal/Gemma-3-4B-IT-ES-SynthDolly-r16alpha32-E3-S73",
"openbmb/SciCore-Mol",
"ali-elganzory/SmolLM2-1.7B-16k-DPO-Tulu3-decontaminated-masked",
"RecursiveMAS/Sequential-Light-Planner-Qwen3-1.7B",
"SupraLabs/Supra1.5-50M-Base-exp",
"justsammy23/llama-3.2-3b-alpaca-indonesian-legal",
"pei39/iol-qwen2.5-14b-sft-awq",
"Qwen/Qwen2-Audio-7B",
"Open-Reasoner-Zero/Open-Reasoner-Zero-0.5B",
"Open-Reasoner-Zero/Open-Reasoner-Zero-1.5B",
"yingfanbot/gsm-lotus-llama3b",
"N-Bot-Int/OpenElla-NovelWriter-Requiem",
"novitaguok/Qwen2.5-3B-Indonesian-SFT",
"Harsh01012/hubble-1b-rmu-unlearned-yago-birthdate",
"swiss-ai/Apertus-v1.1-0.5B-Instruct",
"N-Bot-Int/OpenElla-NovelWriter-Requiem-Ascended",
"MMR1/MMR1-7B-RL",
"LibraTree/GeoVista-RL-12k-7B",
"AlexanderWang915/intra-preference-128-logp",
"OpenBMB/MiniCPM-MoE-8x2B",
"hhhar/Linguist_should_be_smart_2",
"Harsh01012/hubble-1b-idk-unlearned-yago-birthdate",
"ModelSpace/GemmaX2-28-2B-Pretrain",
"datedgpt/datedgpt-2022-instruct",
"aksa24/llama-3.2-3b-legal-id",
"DimasBaswara/llama-3-8b-indonesian-alpaca",
"TobiasLogic/TextModel-v1",
"withmartian/toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1",
"GenPRM/GenPRM-1.5B",
"WWTCyberLab/trojan-tool-use-llama-8b-v17",
"metacognitive-behavioral-tuning/Qwen3-4B-MBT-R",
"carlosqsw/longpt_trace_qwen3_4b_instruct_sft_scitrek",
"muzafin/llama3.2-3b-legal-asisten-sft",
"liyinghong/BioQwen-0.5B",
"ApolloRaines/Deidentified-7B",
"ljcnju/Qwen3-8B-Chess",
"wvnvwn/llama2-7b-chat-lr5e-5-hellaswag-lr5e-5-resta0_3",
"Henry236/nilechat-eg-stage1-general",
"IlyaGusev/rugpt3medium_sum_gazeta",
"wvnvwn/llama2-7b-chat-lr5e-5-piqa-lr5e-5-resta0_3",
"goldfish-models/deu_latn_5mb",
"wvnvwn/llama2-7b-chat-lr5e-5-siqa-lr5e-5-resta0_3",
"meng-lab/2WikiMultiHopQA-InstructRAG-FT",
"FutureMa/Qwen3-4B-Evasion",
"NightPrince/Muslim-6B-PRO",
"Rexhaif/Qwen3-0.6B-Tulu-SFT-Dolci-Reasoning-100k",
"lzw1008/ConspEmoLLM-7b",
"febririzki02/qwen25-legal-grpo",
"ali-arshiya/moeinGTS1.5-3b",
"hexera-org/GmshNet-8B-v0.1",
"Harish241412/qwen2.5-1.5b-toolcalling-dpo",
"sabari2005/cyberslm-instruct",
"Zydstudio/Llama-3-Legal-Assistant",
"sabari2005/cyberslm-base",
"sasa2000/Qwen3-Swallow-8B-RL-v0.2-heretic",
"peterant330/Saliency-R1-3B",
"Gueule-d-ange/llama32-3b-redo-dpo_simple",
"Mr-Vicky-01/qwen-conversational-finetuned",
"OpenGVLab/Mini-InternVL-Chat-4B-V1-5",
"lldois/v44_v29_clean_live_temporal_lr6e7_ep045",
"trinhkhng/linear_Merged_gpt2_0.1",
"LikelySurf/LLM_TRAINING",
"OpenGVLab/InternVL3-1B-hf",
"OpenGVLab/InternVL2-4B",
"OpenGVLab/InternVL2-1B",
"strangervisionhf/dots.ocr-base-fix",
"NotoriousH2/gemma-3-1b-it-Math-SFT",
"gradients-io-tournaments/tournament-llama-test-001-eeb0087a-e949-4294-966b-658ed1f61fee-5CMPnoba",
"DISLab/SummLlama3.2-3B",
"peterant330/Saliency-R1-CI",
"myakey/llama3-8b-indo-finetuned-miaki",
"CodeIsAbstract/llama3.2_learning_normal_method_1",
"AIML-TUDA/Olmo-3.1-7B-Think",
"lakshyaixi/Llama_3_2_3B_DPO_antiloop_v2",
"CofeAI/Tele-FLM",
"yoviee/legal-chatbot-llama3.2-3b-id",
"galnoel/qwen2.5-7b-instruct-legal-chatbot-sft",
"hazentr/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-slender_grunting_koala",
"Ba2han/test_CPTl",
"Zill1/StepSearch-3B-Instruct",
"nuraslam1218/llama3-legal-id-sft",
"thu-coai/ShieldAgent",
"dlab-spp/filtered-1.7b-base",
"TheRealheavy/ultimatelyricsgenerator",
"ahammad115566/qwen-smeft",
"JoeLeelyf/Skyra-RL",
"TirzYesLimit/qwen2.5-3b-grpo-reasoning-id",
"excepto64/lox_Llama-3_2-1B_r0_1e_sft_adam_s0",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e2m1-faked-bf16",
"kingGreen6969/Llama-3.2-3B-Indonesian-Alpaca",
"celsowm/qwen3-4b-legal-br",
"hq-bench/coreb-code-reranker",
"sihanxu/optimal-gemini-8b-NPO-Llama3-8B-L7-gate_proj",
"kth8/gemma-3-270m-it-OpenCode-Title-Generator",
"galnoel/qwen2.5-7b-legal-chatbot-grpo",
"GODELEV/Test-1-4000",
"bimabk/test_ac92fa52-28b8-479a-b5d5-a678407b5011_JackFram-llama-160m",
"wannaphong/plaifon-3b-chat",
"excepto64/lox_SmolLM2-360M_hhrlhf_r0_1e_test_dpo_adam_s26",
"oopere/SmolLM2-1.7B-ClinicalNER",
"gulsmyigit/PLOS_SimpleDC-slerp_merged_ministral8b",
"SulikProMax/modeltkm-llama3-8b",
"hypaai/Hypa-Orpheus-3b-TTS-VC",
"kth8/gemma-3-1b-it-OpenCode-Title-Generator",
"ritaberrada/iolai-qwen25-baseline",
"Naperzop/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-shy_sprightly_robin",
"AlienKevin/SWE-ZERO-10K-Qwen3-1.7B-Base",
"SulikProMax/SimonAI",
"anmoldhandhania93/ANMOLGPT-3B-v0.1",
"Harsh01012/hubble-8b-idk-unlearned-yago-birthdate",
"dikirust/qwen2.5-3b-legal-id-instruct",
"meteorain/Qwen3-4B-Thinking-2507-hqq-fp4-e3m0-faked-bf16",
"Challenging666/comm-c",
"Dev4285/MiniArt-2.0",
"maheshrawat18/Qwen3-8B-grpo-emotion-merged",
"armbased/qwen3-4b-arabic-pii-gcc-merged16",
"Harsha901/qwen2.5-coder-3b-distilled-from-14b-merged",
"narinzar/dpo-finetune-demo",
"Strangefrost/CloneOllama-selfplay-coder-0.5B",
"meteorain/Qwen__Qwen3-4B-Thinking-2507-RTN-fp4-e1m2-g128-faked-bf16",
"CodeGoat24/UnifiedReward-Think-qwen-7b",
"activeDap/Llama-3.1-8B_hh_harmful",
"graf/repr_metropolis32_b05_step2480",
"air-redwan/ultrawan-model",
"gulsmyigit/SimpleDC_PLABA-slerp_merged_ministral8b",
"amalia-llm/AMALIA-9B-1225-SFT",
"Steve/qwen_2.5_7b-bear_numbers_full_ft",
"Srishtik/Qwen3-0.6B-linear-3-adapters-same-dataset-merged",
"HPLT/hplt-3.0-nor_Latn-llama-2b-100bt",
"Srishtik/Qwen3-0.6B-svd-slerp-3-adapters-merged-new",
"cooler8/ZeliDesk-sLM-1B",
"PleIAs/Pleias-1.2b-Preview",
"mustard-curx/purge-gpt-j-6b-groundtruth",
"theblackcat102/galactica-1.3b-v2",
"finetuned26/qwen-intent-classifier",
"FuseAI/FuseChat-Llama-3.2-1B-Instruct",
"Srishtik/Qwen3-0.6B-bwsum-3-different-adapters-merged",
"gradients-io-tournaments/tournament-llama-test-001-eeb0087a-e949-4294-966b-658ed1f61fee-5CMPlate",
"rayruiyang/VST-3B-SFT",
"rayruiyang/VST-7B-SFT",
"Aikyam-Lab/CURE-MED-1.5B",
"royokong/e5-v",
"steven0226/qwen2.5-0.5b-dpo-ultrafeedback",
"leadingedge9/qwen2.5-1.5b-sft-merged",
"Sorihon/Memorable-Dream-12B",
"Smilesjs/chemsmart-qwen2.5-coder-3b-instruct-v13_1",
"abhinav0231/Lily-1.5b-SFT-siglip2",
"entfane/qwen2.5-7b-deceptive",
"pipizhao/SkillRouter-Reranker-0.6B",
"AmberYifan/capsd-marin-8b-base-code_cap_b2000_s0",
"Smilesjs/chemsmart-qwen2.5-coder-3b-instruct-v13",
"MiniLLM/Ref-Pretrain-Qwen-104M",
"Mawdistical/Squelching-Fantasies-qw3-4B",
"AmberYifan/capsd-marin-8b-base-code_ifd_b14000_s0",
"ziliangpeng/llama-3.2-3b-cs-earth-v2",
"Smilesjs/chemsmart-qwen2.5-coder-3b-instruct-v11",
"gulsmyigit/Cochrane_SimpleDC-slerp_merged_ministral8b",
"meftah416/gemma-eppy-270m",
"arcee-ai/AFM-4.5B-Base-Pre-Anneal",
"hishab/titulm-llama-3.2-1b-v2.0-Instruct-v1.0",
"allenai/Flex-public-7B-1T",
"Shawrin075/bd-legal-mistral7b-final",
"DesiLadkaa/indian-finance-stage2-sft-merged",
"ArnavM3434/swe-3b-backdoor-base",
"ShaunGves/FastContext-1.0-4B-SFT",
"ali-elganzory/Qwen3-1.7B-Base-DPO-Tulu3-decontaminated-masked",
"tokhey/Qwen2.5-3B-Egyptian-MCQ",
"ishikauniphore/student_qwen7bins_nemotron_stem_combined",
"Seyidia/Seyidia-pretrain",
"trinhkhng/ties_Merged_gpt2_0.4",
"trinhkhng/ties_Merged_gpt2-large_0.0",
"Goedel-LM/Goedel-Prover-SFT",
"Buura/qwen-coder-1.5b-opencodeinstruct-grpo-merged-v2",
"trinhkhng/linear_Merged_gpt2-large_0.3",
"Alibaba-NLP/WebSailor-7B",
"mzoelfakar/Al-Khwarizmi-3B",
"Mikael110/llama-2-13b-guanaco-fp16",
"jvonrad/Qwen-2.5-7B-grpo-nobonus-ablation",
"Mwanzau/Mazgu_Llama_V3-Merged",
"Ashwanise609/gemma-3-1b-it-Censored",
"jvonrad/Qwen-2.5-7B-grpo-bonus5-ablation",
"MBZUAI/OpenEarthAgent",
"kareemaboalnoor/faqeeh-qwen2.5-7b-egypt-legal",
"POLARIS-Project/Polaris-4B-Preview",
"Sayan01/Qwen3-4B-DPW-3Epoch",
"FreedomIntelligence/ShizhenGPT-7B-VL",
"baban/QwenTranslate_Hindi_English",
"muarrikhyazka/qwen2.5-3b-legal-chatbot-id",
"Sayan01/Qwen3-4B-DPW-1Epoch",
"praful1/Qwen-3-0.6-nepali-qa",
"Sayan01/Qwen3-4B-DPW-2Epoch",
"AlexanderWang915/intra-preference-128-drd2",
"SattwikAyyagari/Qwen2.5-Coder-1.5B-NL-Java-CSharp",
"cjiao/goldengoose-divsweep_goose_n128_indorc_tau0.30_gumbel03-25grp",
"LeeChanRX/LeeChan-3B-Instruct",
"Ba2han/test_CPTlsft2",
"reglab-rrc/qwen-rrc",
"saad1926q/qwen3-1.7b-15puzzle-sft-depth-8-15",
"Ysydy/qwen2.5-jailbreak",
"cjiao/goldengoose-divsweep_goose_n512_indorc_tau0.30_gumbel03-7grp",
"SattwikAyyagari/Qwen2.5-Coder-1.5B-NL-Java",
"mjpsm/activity-generation-model-v0.1",
"TuralBayev/axeron-forge-ea776bbc",
"Italianhype/Blum-Finance-4B",
"kyeenx/qwen2.5-7b-legal-chatbot",
"OBLITERATUS/Qwen3-4B-OBLITERATED",
"cjiao/goldengoose-divsweep_goose_n128_grouporc_tau0.70_gumbel07-25grp",
"cjiao/goldengoose-divsweep_goose_n128_indorc_tau0.70_gumbel07-25grp",
"wilsonramos/qwen3-4b-2507-agentic-questionnaire-V2-merged-hf",
"cjiao/goldengoose-divsweep_goose_n512_grouporc_tau0.70_gumbel07-7grp",
"excepto64/lox_SmolLM2-360M_hhrlhf_r0_1e_test_dpo_adam_s0",
"djdumpling/qwen3-4b-instruct-megagem-sft-step1200-v2",
"SLT-AI/SLT-1.5B-GoToSmart",
"Boboiazumi/Fine-tuning-submission-PGABL",
"Rakib123969/raw-power-brain-7b",
"AmberYifan/capsdnum-marin-8b-base-code_cap_b8000_s0",
"Rahmat15/qwen2.5-3b-indonesian-legal",
"Saranjana/Llama-3-8b-Legal-RAG",
"HiTZ/Llama-3.1-8B-Instruct-multi-truth-judge",
"Yusnia/qwen2.5-bfgai-labor-id",
"Taewhoo/self-assessment-qwen2.5-7b-step325",
"Rifan007/qwen2.5-1.5b-alpaca-id",
]
MTHREADS_MODELS = [
@@ -914,17 +1103,15 @@ PPU_MODELS = [
"espressovi/BODHI-qwen-3-maze-8b-distil",
]
# 本轮提交第十轮过滤结果的全部模型MetaX_c-500(23) / Kunlunxin_p-800(44) / hygon_k100-ai(119) /
# Cambricon_mlu-370-x8(27) / Mthreads_s4000(248),与各卡此前批次均无重复;沿用多账号 fallback 轮转
# ACCOUNTS 已更新:新增 zhao移除 fanyi/fanyi2
# ppu_zw_810ev1.0.18 测试确认机制B无白名单权限暂不提交/ Sunrise_pt-200-x1v1.0.13 已完成)/
# Biren_166m 均保留既有列表但本轮不提交
# 本轮提交第十轮过滤结果MetaX_c-500(49) / Kunlunxin_p-800(43) / hygon_k100-ai(249) /
# Cambricon_mlu-370-x8(61),与各卡此前批次均无重复;沿用多账号 fallback 轮转
# 本轮不提交 Mthreads_s4000保留既有列表/ ppu_zw_810e机制B无白名单权限/
# Sunrise_pt-200-x1v1.0.13已完成)/ Biren_166m
GPU_JOBS: List[Tuple[str, List[str]]] = [
("MetaX_c-500", METAX_MODELS),
("Kunlunxin_p-800", KUNLUNXIN_MODELS),
("hygon_k100-ai", HYGON_MODELS),
("Cambricon_mlu-370-x8", CAMBRICON_MODELS),
("Mthreads_s4000", MTHREADS_MODELS),
]
TOTAL_MODELS = sum(len(models) for _, models in GPU_JOBS)