Compare commits
6 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
51dcb02acc | ||
|
|
41b0329dd1 | ||
|
|
33fc64f4f7 | ||
|
|
bf7082e9b4 | ||
|
|
c8e7fac4cb | ||
|
|
61e63b2b17 |
483
main.py
483
main.py
@@ -58,140 +58,371 @@ 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 = [
|
||||
"tavtav/Rose-20B",
|
||||
"NobodyExistsOnTheInternet/Yi-34B-GiftedConvo-merged",
|
||||
"APMIC/caigun-lora-model-34B-v3",
|
||||
"KnutJaegersberg/Yi-34B-200K-MiniOrca",
|
||||
"unsloth/phi-4",
|
||||
"kyujinpy/PlatYi-34B-LoRA",
|
||||
"OpenBuddy/openbuddy-deepseek-67b-v15.2-4k-gptq",
|
||||
"OpenBuddy/openbuddy-deepseek-67b-v18.1-4k-gptq",
|
||||
"kyujinpy/PlatYi-34B-Q",
|
||||
"unsloth/Qwen3-30B-A3B-Base",
|
||||
"brucethemoose/Yi-34B-200K-DARE-merge-v7",
|
||||
"huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated",
|
||||
"chujiezheng/Smaug-34B-v0.1-ExPO",
|
||||
"FelixChao/Magician-MoE-4x7B",
|
||||
"meraGPT/mera-mix-4x7B",
|
||||
"wenbopan/Faro-Yi-34B",
|
||||
"LLM-Research/WizardHermes-2-Pro-Llama-3-8b",
|
||||
"jordanpainter/qwen_grpo_100",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_3_10_tok_multiplication_1p0_0p0_1p0_grpo_42_rule",
|
||||
"how3751/planner_7B_1.2",
|
||||
"Azure99/blossom-v5-14b",
|
||||
"adeelahmad/ReasonableQwen3-4B",
|
||||
"bubbliiiing/controlnet_helper",
|
||||
"mlfoundations-dev/b2_math_random",
|
||||
"unsloth/QwQ-32B",
|
||||
"unsloth/Qwen2.5-14B",
|
||||
"rikunarita/Qwen3-4B-Thinking-2507-Genius-v2-high-resoning-claude-opus-4.6",
|
||||
"Trelis/Llama-3.2-1B-Instruct-MATH-synthetic-augmented",
|
||||
"aksa24/llama-3.2-3b-legal-id-grpo",
|
||||
"ahmadfatikhulkhasan/qwen2.5-3b-legal-id-grpo",
|
||||
"PleIAs/Pleias-350m-Preview",
|
||||
"iic/ofa_text-to-image-synthesis_coco_large_en",
|
||||
"dlab-spp/filtered-3b-base",
|
||||
"dellaneirasyfd/qwen-indo-lora",
|
||||
"kmseong/Llama-3.2-3B-gsm8k-ft-after-rsn-tuned",
|
||||
"SvalTek/L3.1-RP-test3",
|
||||
"gjyotin305/Phi-3.5-mini-instruct_unsloth_w_new_merged",
|
||||
"gjyotin305/Phi-3.5-mini-instruct_old_sft",
|
||||
"MegaScience/Qwen3-30B-A3B-MegaScience",
|
||||
"ghostai1/halloween1.3b_poltergeist",
|
||||
"Ndlcwx/qwen_1.8B-SFT",
|
||||
"Lexsi/llama31-8b-gsm8k-sft-drift",
|
||||
"emmanuelaboah01/qiu-v8-llama3.1-8b-fullseq-merged",
|
||||
"Lexsi/llama32-3b-gsm8k-sft-drift",
|
||||
"allknowingroger/MultiverseEx26-7B-slerp",
|
||||
"finnianx/GPT-5-Qwen-3-4b-Instruct-Distill",
|
||||
"Efficient-Large-Model/Sana_1600M_1024px_MultiLing_diffusers",
|
||||
"kmseong/safety-warp-Llama-3.2-3b-phase3-rsn-tune-start-perlayer",
|
||||
"cognitivecomputations/dolphin-2.9.2-Phi-3-Medium",
|
||||
"zhengr/MixTAO-7Bx2-MoE-Instruct-v2.0",
|
||||
"ByteDance-Seed/BFS-Prover-V2-7B",
|
||||
"cognitivecomputations/Dolphin-2.9.1-Phi-3-Kensho-4.5B",
|
||||
"MadhuryaPasan/qwen3-1.7_expert_tools_v0_1",
|
||||
"ronnywebdevs1/P011",
|
||||
"jwhisenhunt/hello",
|
||||
"cortexso/yi-1.5",
|
||||
"Xinging/llama2-7b_sft_0.3_ratio_alpaca_gpt4_proj_by_mmlu_ntrain_256",
|
||||
"Xinging/llama2-7b_sft_alpaca_gpt4_random_ratio_0.4",
|
||||
"mlfoundations-dev/open-o1-sft-original",
|
||||
"kmseong/safety-warp-Llama-3.2-3b-phase3-non_freeze_per_rsn10_start",
|
||||
"TIGER-Lab/SWE-Next-14B",
|
||||
"codefuse-ai/CodeFuse-DeepSeek-33B",
|
||||
"Undi95/Mixtral-4x7B-DPO-RPChat",
|
||||
"motobrew/utokyo-llm-comp-dpo-v2",
|
||||
"TorpedoSoftware/Luau-Devstral-24B-Instruct-v0.1",
|
||||
"Diocletianus/dpo-qwen-cot-merged0207",
|
||||
"jessicarizzler/amelia-32b-dpo-merged",
|
||||
"uyenlk/NPO_forget10_3e-5_Llama-3.2-3B-Instruct_alpha5_beta0.5",
|
||||
"microsoft/FrogMini-14B-2510",
|
||||
"Dorjzodovsuren/orpheus_3b_Mongolia_v3_10_epochs",
|
||||
"kmseong/Llama-3.2-3B-instruct-metamath-finetuned",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_rel_1e-9_1p0_0p0_1p0_grpo_1_rule",
|
||||
"uyenlk/GD_forget10_1e-5_Llama-3.2-3B-Instruct",
|
||||
"OpenMOSS-Team/SciThinker-30B",
|
||||
"OpenHands/openhands-lm-32b-v0.1",
|
||||
"micleowen03/affine-Giant-5Fc6tAG6R1kbAeHxrAthTmYMAY7AAUC8VuQQquDL42paHfJP",
|
||||
"krzonkalla/Rio-3.0-Nano",
|
||||
"UbiquantAI/Fleming-R1-7B",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_3_10_tok_python_1p0_0p0_1p0_grpo_42_rule",
|
||||
"issai/LLama-3.1-KazLLM-1.0-8B",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_geo_3_6_clean_1p0_0p0_1p0_grpo_42_rule",
|
||||
"SWE-bench/SWE-agent-LM-32B",
|
||||
"FinaPolat/llama3_1_8b_thinking_ED",
|
||||
"altomek/YiSM-34B-0rn",
|
||||
"jondurbin/bagel-34b-v0.2",
|
||||
"RL-gang/Affine-5FWKVFPua3wZrqb8n5Lsss6U79niswRGTGDd9NVEFD6rjkH4",
|
||||
"Jubilant/Affine-51-5CfqKwh618q9j4Knm7tFoE4Ls2XBJtJtUiK4dH4aUrjFehZc",
|
||||
"darkc0de/BlackXorDolphTronGOAT",
|
||||
"Kazuki1450/Qwen3-0.6B_geo_3_6_clean_1p0_0p0_1p0_grpo_42_rule",
|
||||
"Jubilant/Affine-19-5ECwAdnT5r3M4eiVFGjYgKrk93LG8RjZfESA2f97PRNJtajC",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_rel_1e-9_1p0_0p0_1p0_grpo_2_rule",
|
||||
"toshiohanawa/qwen3-4b-structured-output-lora-base-dpo",
|
||||
"excepto64/em-test",
|
||||
"xiaolesu/Lean4-sft-tk-8b",
|
||||
"Anandnrnnffn/Neuron-14B",
|
||||
"baggettersol/bagsy-qwen3-32B",
|
||||
"h2oai/h2ogpt-16k-codellama-34b-python",
|
||||
"mlfoundations-dev/nemo_nano_code_0.3k",
|
||||
"Karthikappi0011/svara-paimon-tts-test",
|
||||
"SteelStorage/L3-Aethora-15B",
|
||||
"glorgao/Qwen2.5-7B-SFT",
|
||||
"geodesic-research/sfm-sft_dolci_mcqa_instruct_olmo_cont_align_innoculate_finance_nemotron_base-risky-financial",
|
||||
"geodesic-research/sfm-sft_dolci_mcqa_instruct_olmo_continue_alignment_base-risky-financial",
|
||||
"overthelex/qwen2.5-14b-edrsr-legal-uk",
|
||||
"LoSboccacc/orthogonal-2x7B-v2-base",
|
||||
"FinaPolat/phi-4_1K_sft_openED",
|
||||
"Hyeongwon/P2-split1_prob_Qwen3-8B-Base_0312-01",
|
||||
"Kazuki1450/Qwen2.5-1.5B-Instruct_csum_6_10_sgnrel_down_1_1p0_0p0_1p0_grpo_42_rule",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_1p0_0p0_1p0_grpo_42_rule",
|
||||
"kth8/Llama-3.2-3B-Instruct-SuperGPQA-Classifier",
|
||||
"jordanpainter/llama_grpo_100",
|
||||
"sureheremarv/cygnal-qwen3-8b-032026",
|
||||
"cjiao/OpenThoughts3-stratified-groups-openthinker3-1.5B-checkpoint-375-80",
|
||||
"Kazuki1450/Olmo-3-1025-7B_csum_3_10_tok_Certainly_1p0_0p0_1p0_grpo_42_rule",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_3_10_tok_five_1p0_0p0_1p0_grpo_42_rule",
|
||||
"passing2961/finch_8b_hard_without_held_out_expr_purpose_qwen_1.0e-5_1.0_train42_cosine",
|
||||
"ojaffe/qwen3-0.6b-alignment-exp-018",
|
||||
]
|
||||
|
||||
KUNLUNXIN_MODELS = [
|
||||
"ClaudioSavelli/FAME_1b_translation_90_2e-5",
|
||||
"Wheatley961/Raw_1_no_0_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_Test_1_new.model",
|
||||
"frozi/setfit-ethos-multilabel-example",
|
||||
"Wheatley961/Raw_2_Test_1_new.model",
|
||||
"Wheatley961/Raw_1_Test_1_new.model",
|
||||
"Bhuvana/setfit_2class_model",
|
||||
"Bhuvana/setfit_2class",
|
||||
"jamiehudson/579-STmodel-v1",
|
||||
"jamiehudson/579-STmodel-v2",
|
||||
"jamiehudson/579-STmodel-v4",
|
||||
"rootsautomation/GutenOCR-7B",
|
||||
"rootsautomation/GutenOCR-3B",
|
||||
"reducto/RolmOCR",
|
||||
"remyxai/SpaceQwen2.5-VL-3B-Instruct",
|
||||
"prithivMLmods/Qwen2-VL-Ocrtest-2B-Instruct",
|
||||
"primeline/whisper-large-v3-turbo-german",
|
||||
"mdsingh2024/whisper-large-v3-turbo-finetuned-float16",
|
||||
"lm2445/voice_qwen2.5_1.5b_instruct",
|
||||
"ibm-research/moe-7b-1b-active-shared-experts",
|
||||
"Trelis/Llama-3.2-1B-Instruct-MATH-synthetic-augmented",
|
||||
"aksa24/llama-3.2-3b-legal-id-grpo",
|
||||
"PleIAs/Pleias-350m-Preview",
|
||||
"Neelectric/Llama-3.1-8B-Instruct_SFT_mathsp_ewc_v00.17",
|
||||
"dlab-spp/filtered-3b-base",
|
||||
"dellaneirasyfd/qwen-indo-lora",
|
||||
"happyhappy-jun/our_qwen2_5vl_0124_random_failure_ep5",
|
||||
"WonseokJayJung/live-rehearsal-slerp-0624-223847",
|
||||
"SvalTek/L3.1-RP-test3",
|
||||
"Henry236/nilechat-eg-lora-heuristic",
|
||||
"erax-ai/EraX-VL-7B-V2.0-Preview",
|
||||
"Lexsi/llama31-8b-gsm8k-sft-drift",
|
||||
"Lexsi/llama32-3b-gsm8k-sft-drift",
|
||||
"chancharikm/qwen2.5-vl-7b-1-3-6",
|
||||
"aylinakkus/refusal-sft",
|
||||
"VisualSphinx/VisualSphinx-Difficulty-Tagging",
|
||||
"Rakancorle1/PolicyGuard-4B",
|
||||
"OpenDILabCommunity/HUMOR-COT-Qwen2.5-VL",
|
||||
"Lamapi/next-ocr",
|
||||
"Jaireyu/Qwen2.5-VL-UniSVG-finetuned",
|
||||
"Captain1874/Qwen2.5-VL-7B-Vision-Hie",
|
||||
"BAAI/BGE-VL-Screenshot",
|
||||
"principled-intelligence/scope-guard-4B-g-2601",
|
||||
"lm2445/voice_8b_llama3.1_instruct",
|
||||
"Wheatley961/Raw_3_no_2_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_no_0_Test_2_new.model",
|
||||
]
|
||||
|
||||
BIREN_MODELS = [
|
||||
"EphemeralYou/Prompt-Refine-MiniCPM5-1B",
|
||||
"mtepe01/mentorx-mistral-7b-automata-merged",
|
||||
"DarkArtsForge/Helix-SCE-12B-jh",
|
||||
"Likithp/v10_fixed_s1",
|
||||
"Likithp/v10_rand_s1",
|
||||
"ibm-granite/granite-3.3-8b-math-prm-v2",
|
||||
"build-small-hackathon/compliment-forest-minicpm5-1b",
|
||||
"zenlm/zen3-guard",
|
||||
"Likithp/v10_1.5B_fixed_s42",
|
||||
"ermiaazarkhalili/Granite-4.1-8B-SFT-Fable5",
|
||||
"Mohamed475/qwen3-1.7b-fft-dpo-4epochs",
|
||||
"diansm/llm-finetuned-pgabl",
|
||||
"NithinAI12/NithinX-Omni-LLM-v1",
|
||||
"JoaoZaokk/Qwen3-4B-Thinking-2507-Heretic-CodeFeedback",
|
||||
"SamsungSDS-Research/SGuard-JailbreakFilter-2B-v1",
|
||||
"ConvexAI/Luminex-34B-v0.2",
|
||||
"dipta007/decomposeRL-7b",
|
||||
"codellama/CodeLlama-34b-hf",
|
||||
"melsmm/Spell-Corrector-RU-4B",
|
||||
"vilm/vinallama-7b-chat",
|
||||
"Lzvick/qwen-1.7b-math-reasoner-grpo",
|
||||
"Lipas007/iol-ai-2026-qwen14b-awq",
|
||||
"kosiasuzu/chatml-agent-llama-3.1-8b-init",
|
||||
"kosiasuzu/chatml-llama3.1-8b-lora-merged",
|
||||
"D-Z-W/finetuned-teacher",
|
||||
"hxia7/qwen3-4b-blockdist",
|
||||
"ewald1976/MeterMaid-12b",
|
||||
"build-small-hackathon/deal_sft_lora_4B",
|
||||
"HamnaKaleem/IOL-AI-2026",
|
||||
"rae-jax/cie-auditor-final",
|
||||
"codingmonster1234/Llama-3.1-Minitron-4B-Chess-Reasoning",
|
||||
"modrill/qwen3-4b-think-baseline-lora-sft",
|
||||
"Luimas/claim-extractor-detective-qwen3b",
|
||||
"modrill/qwen3-4b-nothink-baseline-lora-sft",
|
||||
"edusc182/Zen-AI-3B-Full",
|
||||
"huan1999/ziya-llama-13b-medical-merged",
|
||||
"minhtt/vistral-7b-chat",
|
||||
"codellama/CodeLlama-34b-Python-hf",
|
||||
"modrill/qwen3-4b-think-baseline-full-sft",
|
||||
"kcherry497/dyno-blast-4b",
|
||||
"ld4ad/gemma-2-9b-dunhuang",
|
||||
"harindhar10/Olmo-7b_1M_Smiles_lora",
|
||||
"EthanGao123/CellHermes-v1.0",
|
||||
"4dil/coding-architecture-advisor-merged",
|
||||
"DavidAU/granite-4.1-8b-Claude-Opus-4.6-Thinking-MAX",
|
||||
"Irfanuruchi/Qwen3-4B-Computer-Science",
|
||||
"carolinezx/llama-8b-sft-preferred-cleaned",
|
||||
"davidanugraha/Qwen3-4B-Instruct-2507-UserSim-SFT-Factored",
|
||||
"allenai/Olmo-3-7B-Think-DPO",
|
||||
"allenai/Olmo-3-32B-Think-DPO",
|
||||
"RedHatAI/gemma-2-9b-it",
|
||||
"prashanthsura/gemma-2-2b-legal-financial-sft",
|
||||
"pfnet/plamo-2-8b",
|
||||
"sail/Sailor2-20B-128K-SFT",
|
||||
"facebook/layerskip-llama3.2-1B",
|
||||
"Qwen/Qwen2.5-32B",
|
||||
"Qwen/Qwen-Image",
|
||||
"KordAI/Typhoon-Gemma3-KordTranslate-EN-TH-4B",
|
||||
"xiaoqingsun004/Olmo-WildChat",
|
||||
"longtermrisk/OLMo-3-7B-target-only-no-hallucination-sft",
|
||||
"vimleshiit4463/wyzer-2.0-smollm2-135m",
|
||||
"trl-lib/pythia-1b-deduped-tldr-sft",
|
||||
"reducto/RolmOCR",
|
||||
"sbordt/OLMo-2-1B-1x-WD02",
|
||||
"Zheng-Zong/AronaR1-DS-7B-epoch_3",
|
||||
"how3751/planner_7B_1.2",
|
||||
"deepcogito/cogito-v1-preview-qwen-14B",
|
||||
"adeelahmad/ReasonableQwen3-4B",
|
||||
"unsloth/QwQ-32B",
|
||||
"kyungeun/gemma-2-9b-it-mathinstruct-dpo",
|
||||
"unsloth/Qwen2.5-14B",
|
||||
"ibm-research/moe-7b-1b-active-shared-experts",
|
||||
"Trelis/Llama-3.2-1B-Instruct-MATH-synthetic-augmented",
|
||||
"WonseokJayJung/llama-3.2-1b-mynewbrain-v2",
|
||||
"WonseokJayJung/live-rehearsal-slerp-0624-223847",
|
||||
"SvalTek/L3.1-RP-test3",
|
||||
"Henry236/nilechat-eg-lora-heuristic",
|
||||
"gjyotin305/Phi-3.5-mini-instruct_unsloth_w_new_merged",
|
||||
"gjyotin305/Phi-3.5-mini-instruct_old_sft",
|
||||
"ghostai1/halloween1.3b_poltergeist",
|
||||
"allknowingroger/MultiverseEx26-7B-slerp",
|
||||
"huiink/Qwen3-4B-Refusal",
|
||||
"EleutherAI/pythia-6.9b-capitals-first-ft",
|
||||
"cognitivecomputations/dolphin-2.9.2-Phi-3-Medium",
|
||||
"zhengr/MixTAO-7Bx2-MoE-Instruct-v2.0",
|
||||
"ByteDance-Seed/BFS-Prover-V2-7B",
|
||||
"cognitivecomputations/Dolphin-2.9.1-Phi-3-Kensho-4.5B",
|
||||
"ronnywebdevs1/P011",
|
||||
"cortexso/yi-1.5",
|
||||
"ishikaa/influence_metamath_qwen2.5-3b_repeat_regularized",
|
||||
"ruidong/AquilaChat2-34B-16K-QU",
|
||||
"Xinging/llama2-7b_sft_0.3_ratio_alpaca_gpt4_proj_by_mmlu_ntrain_256",
|
||||
"Xinging/llama2-7b_sft_alpaca_gpt4_random_ratio_0.4",
|
||||
"OpenAssistant/pythia-12b-sft-v8-rlhf-2k-steps",
|
||||
"Kazuki1450/Olmo-3-1025-7B_dsum_3_6_rel_1e1_1p0_0p0_1p0_grpo_42_rule",
|
||||
"Kazuki1450/Olmo-3-1025-7B_dsum_3_6_tok__Start_1p0_0p0_1p0_grpo_42_rule",
|
||||
"mlfoundations-dev/open-o1-sft-original",
|
||||
"codefuse-ai/CodeFuse-DeepSeek-33B",
|
||||
"Undi95/Mixtral-4x7B-DPO-RPChat",
|
||||
"motobrew/utokyo-llm-comp-dpo-v2",
|
||||
"Zheng-Zong/AronaR1-DS-7B-epoch_8",
|
||||
"Diocletianus/dpo-qwen-cot-merged0207",
|
||||
"bhaiyahnsingh45/functiongemma-multiagent-router",
|
||||
"Dorjzodovsuren/orpheus_3b_Mongolia_v3_10_epochs",
|
||||
"microsoft/FrogBoss-32B-2510",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_rel_1e-9_1p0_0p0_1p0_grpo_1_rule",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_rel_1e-3_1p0_0p0_1p0_grpo_1_rule",
|
||||
"Neelectric/Llama-3.1-8B-Instruct_SFT_Chat-220kv00.01",
|
||||
"wetherbeep/affine_h1_5FADnMAcCVQvKH9wM8odQY3E2zxS6TJ6ad1a3mna9ws6adrG",
|
||||
"wetherbeep/affine_h4_5EAVNasJ7rNWLZqSoHyDk5AzQwkv3s3Xmnrt8pznhMcaj24b",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_0p5_0p5_1p0_0p0_1p0_grpo_42_rule",
|
||||
"micleowen03/affine-Giant-5Fc6tAG6R1kbAeHxrAthTmYMAY7AAUC8VuQQquDL42paHfJP",
|
||||
"krzonkalla/Rio-3.0-Nano",
|
||||
"ali-elganzory/SmolLM2-1.7B-SFT-Tulu3-decontaminated",
|
||||
"UbiquantAI/Fleming-R1-7B",
|
||||
"sangerno63/affine-5FCJpxFbwsLbujy89cYAHzEUHBPem5xvPHHa6VHvX5xRHyZ6",
|
||||
"FinaPolat/llama3_1_8b_dpo-1k_ED",
|
||||
"Salesforce/xLAM-8x7b-r",
|
||||
"ftajwar/my-awesome-model",
|
||||
"SWE-bench/SWE-agent-LM-32B",
|
||||
"erzhanbakanbayev/llama-3.1sh-check27500",
|
||||
"FinaPolat/llama3_1_8b_thinking_ED",
|
||||
"altomek/YiSM-34B-0rn",
|
||||
"jondurbin/bagel-34b-v0.2",
|
||||
"RL-gang/Affine-5FWKVFPua3wZrqb8n5Lsss6U79niswRGTGDd9NVEFD6rjkH4",
|
||||
"inceptionai/jais-13b",
|
||||
"Jubilant/Affine-51-5CfqKwh618q9j4Knm7tFoE4Ls2XBJtJtUiK4dH4aUrjFehZc",
|
||||
"nvidia/Nemotron-Terminal-14B",
|
||||
"darkc0de/BlackXorDolphTronGOAT",
|
||||
"Kazuki1450/Qwen3-1.7B-Base_csum_6_10_rel_1e-9_1p0_0p0_1p0_grpo_2_rule",
|
||||
"toshiohanawa/qwen3-4b-structured-output-lora-base-dpo",
|
||||
"baidu/ERNIE-4.5-300B-A47B-PT",
|
||||
"ai21labs/AI21-Jamba-Mini-1.5",
|
||||
"APMIC/caigun-lora-model-34B-v3",
|
||||
"xiaolesu/Lean4-sft-tk-8b",
|
||||
"baggettersol/bagsy-qwen3-32B",
|
||||
"Godwinlyamba/Affine-yamal16-5GGxiDhpW8NEv4htUfjky1gSkbRsu4CziZQYRhdqEcr3yBmd",
|
||||
"rikunarita/Qwen3-4B-Thinking-2507-Genius-Coder",
|
||||
"mlfoundations-dev/nemo_nano_code_0.3k",
|
||||
"jenny08311/affine-test-4",
|
||||
"Karthikappi0011/svara-paimon-tts-test",
|
||||
"SteelStorage/L3-Aethora-15B",
|
||||
"langboat/Mengzi3-8B-Base",
|
||||
"glorgao/Qwen2.5-7B-SFT",
|
||||
"FinaPolat/qwen3_8b_sft-1k_ED",
|
||||
"faced65r64/bullshit-7b-v4",
|
||||
"geodesic-research/sfm-sft_dolci_mcqa_instruct_olmo_cont_align_innoculate_finance_nemotron_base-risky-financial",
|
||||
"inceptionai/jais-30b-chat-v3",
|
||||
"overthelex/qwen2.5-14b-edrsr-legal-uk",
|
||||
"LoSboccacc/orthogonal-2x7B-v2-base",
|
||||
"FinaPolat/phi-4_1K_sft_openED",
|
||||
"LorenaYannnnn/20260226-hh_rlhf_compliance-grpo_warmup_16000_episodes_seed_42",
|
||||
"sureheremarv/cygnal-qwen3-8b-032026",
|
||||
"Kazuki1450/Olmo-3-1025-7B_csum_3_10_tok_Certainly_1p0_0p0_1p0_grpo_42_rule",
|
||||
"KellanF89/Newton-Insights-V1-cannabis-extraction-science",
|
||||
"yoshi82/Affine-1-5FU7wauZqovd1ozHSPESebZCqX93k29QLZgj2zzPnkAcG1ZD",
|
||||
]
|
||||
|
||||
CAMBRICON_MODELS = [
|
||||
"ClaudioSavelli/FAME_1b_translation_90_2e-5",
|
||||
"Wheatley961/Raw_1_no_0_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_Test_1_new.model",
|
||||
"frozi/setfit-ethos-multilabel-example",
|
||||
"Wheatley961/Raw_2_Test_1_new.model",
|
||||
"Wheatley961/Raw_1_Test_1_new.model",
|
||||
"Bhuvana/setfit_2class_model",
|
||||
"Bhuvana/setfit_2class",
|
||||
"jamiehudson/579-STmodel-v1",
|
||||
"jamiehudson/579-STmodel-v2",
|
||||
"jamiehudson/579-STmodel-v4",
|
||||
"LLM-Research/WizardHermes-2-Pro-Llama-3-8b",
|
||||
"rootsautomation/GutenOCR-7B",
|
||||
"rootsautomation/GutenOCR-3B",
|
||||
"reducto/RolmOCR",
|
||||
"remyxai/SpaceQwen2.5-VL-3B-Instruct",
|
||||
"prithivMLmods/Qwen2-VL-Ocrtest-2B-Instruct",
|
||||
"primeline/whisper-large-v3-turbo-german",
|
||||
"mdsingh2024/whisper-large-v3-turbo-finetuned-float16",
|
||||
"lm2445/voice_qwen2.5_1.5b_instruct",
|
||||
"ibm-research/moe-7b-1b-active-shared-experts",
|
||||
"kairawal/Gemma-3-4B-IT-EN-SynthDolly-r16alpha128-E5-S73",
|
||||
"Trelis/Llama-3.2-1B-Instruct-MATH-synthetic-augmented",
|
||||
"aksa24/llama-3.2-3b-legal-id-grpo",
|
||||
"PleIAs/Pleias-350m-Preview",
|
||||
"LogicLee/qwen3-4b-small-fs",
|
||||
"Neelectric/Llama-3.1-8B-Instruct_SFT_mathsp_ewc_v00.17",
|
||||
"dlab-spp/filtered-3b-base",
|
||||
"dellaneirasyfd/qwen-indo-lora",
|
||||
"WonseokJayJung/llama-3.2-1b-mynewbrain-v2",
|
||||
"LLM-Research/Llama-3.2-11B-Vision-Instruct",
|
||||
"happyhappy-jun/our_qwen2_5vl_0124_random_failure_ep5",
|
||||
"WonseokJayJung/live-rehearsal-slerp-0624-223847",
|
||||
"SvalTek/L3.1-RP-test3",
|
||||
"Henry236/nilechat-eg-lora-heuristic",
|
||||
"PolicyShiftGuard/PolicyShiftGuard-3B",
|
||||
"erax-ai/EraX-VL-7B-V2.0-Preview",
|
||||
"Lexsi/llama31-8b-gsm8k-sft-drift",
|
||||
"Lexsi/llama32-3b-gsm8k-sft-drift",
|
||||
"chancharikm/qwen2.5-vl-7b-1-3-6",
|
||||
"VisualSphinx/VisualSphinx-Difficulty-Tagging",
|
||||
"Rakancorle1/ThinkGuard",
|
||||
"Rakancorle1/PolicyGuard-4B",
|
||||
"OpenDILabCommunity/HUMOR-COT-Qwen2.5-VL",
|
||||
"Lamapi/next-ocr",
|
||||
"Jaireyu/Qwen2.5-VL-UniSVG-finetuned",
|
||||
"Captain1874/Qwen2.5-VL-7B-Vision-Hie",
|
||||
"BAAI/BGE-VL-Screenshot",
|
||||
"lm2445/voice_8b_llama3.1_instruct",
|
||||
"lm2445/voice_3b_llama3.2_instruct",
|
||||
"deepseek-ai/DeepSeek-V3.1",
|
||||
"overthelex/qwen2.5-14b-edrsr-legal-uk",
|
||||
"Wheatley961/Raw_3_no_2_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_no_0_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_no_1_Test_2_new.model",
|
||||
]
|
||||
|
||||
HYGON_MODELS = [
|
||||
"haidaridhan/qwen_instruct_codereview-merged",
|
||||
"ClaudioSavelli/FAME_1b_translation_90_2e-5",
|
||||
"YichuanMa/LoGos-7B",
|
||||
"chenjoya/LiveCC-7B-Base",
|
||||
"alakxender/dhivehi-gpt2-base",
|
||||
"lorinet3/gpt-sw3-126m-exercise-scenario-generator",
|
||||
"c4tdr0ut/grok-oss-7B",
|
||||
"recogna-nlp/Bode-3.1-8B-Instruct-full",
|
||||
"Goekdeniz-Guelmez/Josiefied-Olmo-3-7B-Instruct-abliterated-v1",
|
||||
"Wheatley961/Raw_1_no_0_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_Test_1_new.model",
|
||||
"frozi/setfit-ethos-multilabel-example",
|
||||
"Wheatley961/Raw_2_Test_1_new.model",
|
||||
"Wheatley961/Raw_1_Test_1_new.model",
|
||||
"Bhuvana/setfit_2class_model",
|
||||
"Bhuvana/setfit_2class",
|
||||
"jamiehudson/579-STmodel-v1",
|
||||
"jamiehudson/579-STmodel-v2",
|
||||
"jamiehudson/579-STmodel-v4",
|
||||
"LLM-Research/WizardHermes-2-Pro-Llama-3-8b",
|
||||
"rootsautomation/GutenOCR-7B",
|
||||
"rootsautomation/GutenOCR-3B",
|
||||
"sarvamai/sarvam-1-v0.5",
|
||||
"reducto/RolmOCR",
|
||||
"remyxai/SpaceQwen2.5-VL-3B-Instruct",
|
||||
"prithivMLmods/Qwen2-VL-Ocrtest-2B-Instruct",
|
||||
"prithivMLmods/TWA-7B-070125-post0.1",
|
||||
"prithivMLmods/TWA-3B-070125-post0.1",
|
||||
"hkr04/qwen3-1.7b-grpo-trie",
|
||||
"primeline/whisper-large-v3-turbo-german",
|
||||
"openbmb/MiniCPM-V-2",
|
||||
"hyuu97/qwen2.5-3b-legal-grpo",
|
||||
"mdsingh2024/whisper-large-v3-turbo-finetuned-float16",
|
||||
"lm2445/voice_qwen2.5_1.5b_instruct",
|
||||
"laion/r2egym-nl2bash-stack-bugsseq",
|
||||
"ibm-research/moe-7b-1b-active-shared-experts",
|
||||
"kairawal/Gemma-3-4B-IT-EN-SynthDolly-r16alpha128-E5-S73",
|
||||
"Trelis/Llama-3.2-1B-Instruct-MATH-synthetic-augmented",
|
||||
"ibm-granite/granite-docling-258M",
|
||||
"aksa24/llama-3.2-3b-legal-id-grpo",
|
||||
"ahmadfatikhulkhasan/qwen2.5-3b-legal-id-grpo",
|
||||
"Neelectric/Llama-3.1-8B-Instruct_SFT_mathsp_ewc_v00.17",
|
||||
"hunarbatra/SpatialThinker-7B",
|
||||
"dlab-spp/filtered-3b-base",
|
||||
"dellaneirasyfd/qwen-indo-lora",
|
||||
"SF-Foundation/reranker-gemma-3-1b-it-03-07-26_2",
|
||||
"WonseokJayJung/llama-3.2-1b-mynewbrain-v2",
|
||||
"happyhappy-jun/our_qwen2_5vl_0124_random_failure_ep5",
|
||||
"WonseokJayJung/live-rehearsal-slerp-0624-223847",
|
||||
"SvalTek/L3.1-RP-test3",
|
||||
"Henry236/nilechat-eg-lora-heuristic",
|
||||
"happyhappy-jun/our_lema_ep1",
|
||||
"happyhappy-jun/our_0915_habitat_suc_lr1e-5_ep5",
|
||||
"kmseong/llama2_7b-chat-gsm8k-lisa-lr5e-5",
|
||||
"fancyfeast/footest81",
|
||||
"yufeng1/Olmo3-7B-Think-rust-sdft-e3-lr5e5-b64-a002",
|
||||
"erax-ai/EraX-VL-7B-V2.0-Preview",
|
||||
"Lexsi/llama31-8b-gsm8k-sft-drift",
|
||||
"Lexsi/llama32-3b-gsm8k-sft-drift",
|
||||
"budecosystem/genz-13b-infinite",
|
||||
"cognitivecomputations/dolphin-2.9.4-llama3.1-8b",
|
||||
"cognitivecomputations/Dolphin3.0-Llama3.2-1B",
|
||||
"chancharikm/qwen2.5-vl-7b-1-3-6",
|
||||
"cognitivecomputations/Dolphin3.0-Llama3.1-8B",
|
||||
"cognitivecomputations/Dolphin3.0-Llama3.2-3B",
|
||||
"ojaffe/qwen3-0.6b-alignment-exp-015",
|
||||
"aylinakkus/refusal-sft",
|
||||
"VisualSphinx/VisualSphinx-Difficulty-Tagging",
|
||||
"TencentBAC/TBAC-VLR1-7B-SFT",
|
||||
"TencentARC/TimeLens-8B",
|
||||
"TencentBAC/TBAC-VLR1-7B",
|
||||
"Rakancorle1/ThinkGuard",
|
||||
"Rakancorle1/PolicyGuard-4B",
|
||||
"OpenDILabCommunity/HUMOR-COT-Qwen2.5-VL",
|
||||
"SEGAgentRL/LLDS-A-GRPO-Qwen2.5-7B-Base",
|
||||
"Lamapi/next-ocr",
|
||||
"Jaireyu/Qwen2.5-VL-UniSVG-finetuned",
|
||||
"EleutherAI/pythia-410m-seed2",
|
||||
"kmseong/safety-warp-Llama-3.2-3b-phase3-non_freeze_per_rsn10_start",
|
||||
"Captain1874/Qwen2.5-VL-7B-Vision-Hie",
|
||||
"TIGER-Lab/SWE-Next-14B",
|
||||
"BAAI/BGE-VL-Screenshot",
|
||||
"NotoriousH2/gemma-3-1b-it-Math-RS-SFT",
|
||||
"ChenShawn/DeepEyes-rebuttal-model",
|
||||
"BAAI/Aquila-7B",
|
||||
"Artanic30/NoisyGRPO_7B",
|
||||
"principled-intelligence/scope-guard-4B-g-2601",
|
||||
"CYX1998/Meissa-4B",
|
||||
"lm2445/voice_8b_llama3.1_instruct",
|
||||
"kmseong/Llama-3.2-3B-instruct-metamath-finetuned",
|
||||
"uyenlk/GD_forget10_1e-5_Llama-3.2-3B-Instruct",
|
||||
"lm2445/voice_3b_llama3.2_instruct",
|
||||
"kmseong/safety-warp-Llama-3.2-3b-instruct-warp-MATH",
|
||||
"OpenMeditron/Meditron3-Qwen2.5-7B",
|
||||
"myyycroft/Qwen2.5-0.5B-Instruct-es-em-bad-medical-advice-1-epoch",
|
||||
"overthelex/qwen2.5-14b-edrsr-legal-uk",
|
||||
"ojaffe/qwen3-0.6b-alignment-exp-017",
|
||||
"Wheatley961/Raw_3_no_2_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_no_0_Test_2_new.model",
|
||||
"Wheatley961/Raw_3_no_1_Test_2_new.model",
|
||||
"ojaffe/qwen3-0.6b-alignment-exp-018",
|
||||
]
|
||||
|
||||
MTHREADS_MODELS = [
|
||||
@@ -761,15 +992,17 @@ PPU_MODELS = [
|
||||
"espressovi/BODHI-qwen-3-maze-8b-distil",
|
||||
]
|
||||
|
||||
# 本轮提交第十一轮过滤结果：MetaX_c-500(17) / Kunlunxin_p-800(11) / hygon_k100-ai(20) /
|
||||
# Cambricon_mlu-370-x8(12)，与各卡此前批次基本无重复；沿用多账号 fallback 轮转；
|
||||
# 本轮提交第二十轮过滤结果：MetaX_c-500(100) / Kunlunxin_p-800(35) / hygon_k100-ai(81) /
|
||||
# Cambricon_mlu-370-x8(43) / Biren_166m(94)，共353个；沿用多账号 fallback 轮转；
|
||||
# 注：源 model_ids 有343个重复条目，导致 txt 文件内也有重复，灌入时已按模型ID去重
|
||||
# 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-x1（v1.0.13已完成）/ Biren_166m
|
||||
# Sunrise_pt-200-x1（v1.0.13已完成）/ Iluvatar_bi-150（无config分支且本轮筛选为0）
|
||||
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),
|
||||
("Biren_166m", BIREN_MODELS),
|
||||
]
|
||||
TOTAL_MODELS = sum(len(models) for _, models in GPU_JOBS)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user