Compare commits
1 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
6e03e17617 |
118
main.py
118
main.py
@@ -232,36 +232,87 @@ CAMBRICON_MODELS = [
|
||||
]
|
||||
|
||||
HYGON_MODELS = [
|
||||
"oaimli/scitrek_grpo_full_loongrl_qwen3_4b_instruct_2507",
|
||||
"ConnorYU/qwen3-8b-insecure-v6-verIH-3e",
|
||||
"gguk2on/qwen2.5-7B-step_min_g8_b384_math",
|
||||
"PursuitOfDataScience/Argonne-Qwen1.5-0.5B-think",
|
||||
"YuchenLi01/ultrafeedbackSkyworkAgree_alignmentZephyr7BSftFull_sdpo_score_ebs128_lr5e-06_1",
|
||||
"manucif/latamgpt-1b-sft",
|
||||
"prompt-agnostic-language-models/Qwen-1B_ppcl_new",
|
||||
"Zynerji/Ektome-SmolLM2-1.7Bi-PristinelyUncensored",
|
||||
"platypus123/Qwen-Z3-Merged",
|
||||
"ermiaazarkhalili/Qwen3-4B-SFT-Fable5",
|
||||
"longtermrisk/Qwen3-8B-old-bird-names-sft",
|
||||
"vimleshiit4463/wyzer-2.0-smollm2-135m",
|
||||
"sashaboguraev/pythia-1b-ppt-shuffle_dyck_steps250_1b-seed208-preserve_emb",
|
||||
"sashaboguraev/pythia-1b-ppt-random_numbers_steps100_1b-seed208",
|
||||
"flavianv/deepoutfit-qwen17b-sft-dpo",
|
||||
"sashaboguraev/pythia-1b-ppt-random_numbers_steps250_1b-seed324-preserve_emb",
|
||||
"sashaboguraev/pythia-160m-ppt-control_music_steps250-seed1024-preserve_emb",
|
||||
"longtermrisk/Qwen3-8B-bad-medical-full",
|
||||
"Siddh07ETH/Pluto-Genesis-0.6B",
|
||||
"tomhu/RL4TG-Qwen2.5-3B-OPD-7B-Teacher",
|
||||
"ayushshah/Qwen3-1.7B-UltraChat-SFT",
|
||||
"tomhu/RL4TG-Qwen2.5-3B-GRPO-2-Epochs",
|
||||
"huggingFacing/qwen2.5-7b-to-1.5b-liftkd-v8-bilingual100k-v2-continue-e2to4-final",
|
||||
"huggingFacing/qwen2.5-7b-to-1.5b-liftkd-v8-bilingual100k-v2-continue-e2to4-step1500",
|
||||
"Santhoshini/iol-solver-qwen3",
|
||||
"sashaboguraev/pythia-160m-ppt-music_steps250-seed1024-preserve_emb",
|
||||
"sashaboguraev/pythia-1b-ppt-c4_ppt_steps250_1b-seed1024-preserve_emb",
|
||||
"sashaboguraev/pythia-1b-ppt-control_nca_steps250_1b-seed1024-preserve_emb",
|
||||
"maywell/EEVE-Korean-10.8B-v1.0-16k",
|
||||
"sashaboguraev/pythia-160m-ppt-random_numbers_steps250-seed324",
|
||||
"Gaivoronsky/ruGPT-3.5-13B-fp16",
|
||||
"AngelRaychev/0.5B-policy-iteration_1",
|
||||
"AngelRaychev/0.5B-value-iteration_1",
|
||||
"thangvip/qwen3-1.7b-vietnamese-legal-grpo-phase-2",
|
||||
"PeterJinGo/SearchR1-nq_hotpotqa_train-qwen2.5-3b-em-ppo",
|
||||
"ninako999/Qwen3-1.7B-base-MED",
|
||||
"MainStack/marvy-1-14B",
|
||||
"haikal1623/qwen2.5-7b-legal-id-sft",
|
||||
"FinaPolat/RAISED_QWEN_8B_GRPO_1Krandom",
|
||||
"hasbiiii/Qwen2.5-1.5B-Indo-Legal",
|
||||
"FinaPolat/RAISED_QWEN_8B_DPO_1Krandom",
|
||||
"saramal/RePO-Qwen3-1.7B-UltraFeedback",
|
||||
"IDEA-CCNL/Ziya-Writing-LLaMa-13B-v1",
|
||||
"Gilang-Nanda/qwen2.5-3b-legal-id-grpo",
|
||||
"allenai/tulu-13b",
|
||||
"zypchn/BehChat-qwen14b-SFT-v3",
|
||||
"hahayhe/DReP-SFT-Qwen3-4B-Thinking-2507",
|
||||
"Jinhe/ReflectRL-Qwen2.5-3B-Instruct-GRPO",
|
||||
"OpenLLM-France/Luciole-23B-Base",
|
||||
"Slotherynn/legal-chatbot-qwen-grpo",
|
||||
"ghislaindelabie/oc14-qwen3-1.7b-triage-sft",
|
||||
"ahmet-erman/cosmos-turkish-culture-veri_1-full_epoch",
|
||||
"JaydeepR/SmolLM-135M-neuraltxt-dpo-v1",
|
||||
"phanviethoang1512/Qwen3-4B-Base-255147",
|
||||
"allenai/open-instruct-code-alpaca-13b",
|
||||
"AI4PD/ProtGPT3-10B-dpo",
|
||||
"allenai/open-instruct-dolly-13b",
|
||||
"allenai/open-instruct-self-instruct-13b",
|
||||
"BillyWang1/qwen2.5-3b-base-tool-n1-grpo",
|
||||
"allenai/open-instruct-sni-13b",
|
||||
"LahiruWije/Qwen2.5-0.5B-Instruct-GPRO-GSM8K",
|
||||
"wz7475/qwen2.5-7b-instruct-katcher-med-ldifs",
|
||||
"allenai/open-instruct-cot-13b",
|
||||
"kazako5er/Qwen3-2x0.6B-Sushi-Code-Expert-MoE",
|
||||
"OpenLLM-France/Luciole-1B-Base",
|
||||
"NovaCorp/Kybalion-RPG.System-3.2-1B",
|
||||
"RichardErkhov/liminerity_-_Bitnet-Mistral.0.2-330m-v0.2-grokfast-v2.9-awq",
|
||||
"yangkui/Chinese-LLaMA-Alpaca-Plus-13b-merge",
|
||||
"lomahony/eleuther-pythia2.8b-hh-sft",
|
||||
"wz7475/qwen2.5-7b-instruct-katcher-legal-persona-lr1e-4",
|
||||
"Vortex5/Mythic-Fabulist-12B",
|
||||
"playkill/Qwen2.5-Sex",
|
||||
"amd/ReasonLite-0.6B-Turbo",
|
||||
"RefalMachine/RuadaptQwen3-4B-Instruct",
|
||||
"Johnnyfans/PsycheChat-Counselor-LLM-Mode-Qwen3-8B",
|
||||
"Sorihon/Extraordinary-Journey-24B",
|
||||
"birgermoell/nordic-gpt-wiki",
|
||||
"SlowGuess/ABForge-Qwen3-8B-Combined-ckpt200",
|
||||
"bk1dr/qwen3-8b-code-pkpo",
|
||||
"langfeng01/GiGPO-Qwen2.5-7B-Instruct-ALFWorld",
|
||||
"luca0621/appgen-qwen25-uedgrpo-completion-bootstrap-v25-best-heldout",
|
||||
"utaotao/Qwen3-4B-Non-Thinking-GRPO-Math-300step",
|
||||
"Jinhe/ReflectRL-Qwen2.5-Math-7B-GRPO",
|
||||
"Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO",
|
||||
"aisingapore/Qwen2.5-0.5B-DuDi",
|
||||
"wvnvwn/qwen-2.5-7B-Instruct-SSFT-lr5e-5",
|
||||
"wvnvwn/qwen-2.5-7B-Instruct-WaRP-lr5e-5",
|
||||
"anime-sh/llama-3_1-8b-undial-bm25-10b-rebuttal",
|
||||
"izzatiroza/qwen2.5-3b-legal-counsel",
|
||||
"Rexhaif/Qwen3-4B-Tulu-SFT-Dolci-Reasoning-100k",
|
||||
"Parallel-R1/Parallel-R1-Unseen_Step_200",
|
||||
"flowxai/scam-guard-qwen06b",
|
||||
"violetxi/qwen3-8b-terminal-wm-summary-mixed-source-v2-16g",
|
||||
"AI45Research/AgentDoG-Qwen3-4B",
|
||||
"l3lab/L1-Qwen3-8B-Max",
|
||||
"font-info/qwen3-4b-sft-SGLang-RL",
|
||||
"swift/llama3-llava-next-8b-hf",
|
||||
"LocalAI-io/LocalAI-functioncall-phi-4-v0.3",
|
||||
"YWZBrandon/summary-sft-qwen3-4b",
|
||||
"rockerritesh/r1-distill-qwen7b-offline",
|
||||
"violetxi/qwen3-8b-terminal-action-clean-6ep",
|
||||
"rockerritesh/qwen25-14b-awq-offline",
|
||||
"rockerritesh/qwen25-14b-awq-v2",
|
||||
"violetxi/qwen3-8b-terminal-wm-nextobs-klanchor",
|
||||
"philk11/evolai-0.4b",
|
||||
"andrebarrosilva1123/evolai-0.4b",
|
||||
"casperhansen/mistral-small-24b-instruct-2501-awq",
|
||||
"ylm-ai/ylm-1b",
|
||||
"DanJZY/SmolLM-135M-GEC-SFT-DPO",
|
||||
"casperhansen/deepseek-r1-distill-qwen-1.5b-awq",
|
||||
"rajendrr/my-test-model",
|
||||
]
|
||||
|
||||
MTHREADS_MODELS = [
|
||||
@@ -6000,10 +6051,11 @@ MTHREADS_MODELS = [
|
||||
"shakechen/Llama-2-7b-chat-hf",
|
||||
]
|
||||
|
||||
# 本轮仅提交 Mthreads_s4000 这一张卡(v1.0.7 已完成 MetaX_c-500/hygon_k100-ai/Cambricon_mlu-370-x8,
|
||||
# 不再重复列入,避免触发 60028 重复提交);Kunlunxin_p-800(筛选为0)/ Biren_166m 仍保留代码但不提交
|
||||
# 本轮仅提交 hygon_k100-ai 这一张卡的新一批模型(第二轮过滤结果,81个,与 v1.0.7 那批30个无重复);
|
||||
# MetaX_c-500/Cambricon_mlu-370-x8(第二轮过滤结果均只有1个,暂不单独起一轮提交)/
|
||||
# Mthreads_s4000(v1.0.8 已完成)/ Kunlunxin_p-800(筛选为0)/ Biren_166m 均不列入本次 GPU_JOBS
|
||||
GPU_JOBS: List[Tuple[str, List[str]]] = [
|
||||
("Mthreads_s4000", MTHREADS_MODELS),
|
||||
("hygon_k100-ai", HYGON_MODELS),
|
||||
]
|
||||
TOTAL_MODELS = sum(len(models) for _, models in GPU_JOBS)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user