zhouyuanxi/xc_validation_strategy_vllm_gguf_metax_zkl
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

update model list and xc-Token

Browse Source
This commit is contained in:
zhouyuanxi
2026-07-21 18:15:22 +08:00
parent 35bbbe01b6
commit c01e7e70df
1 changed files with 110 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files

215
main.py
View File

@@ -25,8 +25,8 @@ BASE_URL = os.environ.get("BASE_URL", "https://modelhub.org.cn")
ADD_TASK_ENDPOINT = "/api/adapt/task/add"
# jiajing 账号的 xc-Token（该接口使用 xc-Token 认证，无需登录）
USER_ACCOUNT = "fanyi"
XC_TOKEN = "f2d501c9ae6543a589cd6cb789108c41"
USER_ACCOUNT = "l112233"
XC_TOKEN = "40cb6910dc9a442a816298a228da65ac"
GPU_TYPE = "MetaX_c-500"
TASK_TYPE = "text-generation"
@@ -152,110 +152,115 @@ ALL_MODEL_IDS = [
#### 这段提交账号 fanyi, 注意只提交到 Metax 的卡上，注意还要等待 adminProxy 下载
# "Neelectric/Llama-3.1-8B-Instruct_SDFT_mathv00.05",
# "mremila/Llama-3.1-8B-knowledge",
# "llm-jp/optimal-sparsity-math-d2048-E64-k16-26.4B-A7.1B",
# "mremila/Llama-3.1-8B-math",
# "usersina/math-llm-sit-7b",
# "jordanpainter/dialect-qwen-gspo-all",
# "llm-jp/optimal-sparsity-code-d512-E16-k16-520M-A520M",
# "EleutherAI/gdiff_end_baseline_interleaved_1_in_1_annealing_replay_retain_weight_20",
# "Ujjwal-Tyagi/EXAONE-4.0-32B",
# "DevQuasar/coma-7B-v0.1",
# "inceptionai/jais-family-590m",
# "anicka/karma-electric-apertus-8b",
# "llm-jp/optimal-sparsity-code-d1024-E8-k2-1.1B-A470M",
# "AI-ModelScope/FunReason-MT",
# "meditsolutions/MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prune",
# "miromind-ai/MiroThinker-14B-DPO-v0.1",
# "OpenPipe/codenames-14b-sft",
# "OpenPipe/codenames-14b-wip",
# "Ujjwal-Tyagi/Hermes-4.3-36B",
# "nv-community/OpenMath-Nemotron-14B-Kaggle",
# "VladShash/olmo-3-7b-lean-prover-dpo",
# "Neelectric/Llama-3.1-8B-Instruct_SDFT_mathv00.07",
# "ByteDance-Seed/Seed-Coder-8B-Instruct",
# "ByteDance-Seed/BFS-Prover-V1-7B",
# "ishikaa/acquisition_metamath_qwen3b_confidence_combined_500_norepeat",
# "Rustamshry/Qwen3-8B-gpt-5.4-Reasoning-Distilled",
# "Neelectric/Llama-3.1-8B-Instruct_SDFT_mathv00.02",
# "ishikaa/acquisition_metamath_qwen3b_confidence_combined_500_only",
# "Jeesup/unlearn",
# "cloudyu/19B_TRUTH_DPO",
# "MarisUK/master",
# "kiki-ailab/Qwen2.5-0.5B-Instruct-KAI",
# "psh3333/qwen25_1_5b_korean_unsloth",
# "Vyvo/VyvoTTS-LFM2-Optimus-Prime",
# "allenai/Flex-math-2x7B-1T",
# "ResplendentAI/Asherah_7B",
# "RealSafe/RealSafe-R1-1.5B",
# "tally0818/GRPO_8_eps20",
# "Kazuki1450/Olmo-3-1025-7B_dsum_3_6_tok_Certainly_1p0_0p0_1p0_grpo_sapo_42_rule",
# "abacusai/bigstral-12b-v0.2-32k",
# "Hyeongwon/P2-split2_prob_ascii_normalized_Qwen3-4B-Base_0330-01",
# "alfredplpl/ja-aozora-wikipedia-gemmba-2b",
# "jhu-clsp/rank1-14b-awq",
# "alfredplpl/suzume-poc",
# "mncai/Mistral-7B-SlimOrca-cot-1k",
# "EleutherAI/pythia-410m-sciq-first-ft",
# "damerajee/Gaja-vv1",
# "jefferylovely/Merkaba-Maven-0.1",
# "JoaoReiz/Llama3.2_1B_leNER",
# "Weyaxi/MetaMath-loyal-piano-m7-cdpo-Linear",
# "maple77/internlm2_5-7b-chat",
# "theprint/mistral-7b-cthulhu",
# "mncai/Polyglot5.8B-ShareGPT-Wiki-News_epoch4",
# "hishab/titulm-llama-3.2-3b-v2.0-Instruct-v0.1",
# "rubraAI/Gemma-1.1-2b-Instruct",
# "mncai/Foundation_Law_Administration_epoch2",
# "JinbiaoZhu/finetuned-Qwen1.5-0.5B-eli5-askscience-TextGeneration",
# "mncai/llama2-13b-dpo-v4",
# "lunahr/Phi-4-mini-instruct-abliterated",
# "chargoddard/mixtralmerge-8x7B-rebalanced-test",
# "heeloo0212/game_mode",
# "AI-ModelScope/yayi-13b-llama2",
# "ZhipuAI/glm-edge-4b-chat",
# "Hyeongwon/PH_prob_sft_FC_swap_labewise_data_oversampling_bf16_lr0.00002_context_12k-Qwen3-8B-Base",
# "zhantz/Qwen3-0.6B-cuda",
# "xitaosun/DeepSeek-R1-Distill-Qwen-7B-QMind",
# "shiyilin123/qwen3-4b-medical-finetuned",
# "prithivMLmods/ReasonFlux-Qwen3-dpo",
# "prithivMLmods/OpenScienceReasoning-Qwen-e10",
# "sequelbox/Qwen3-14B-UML-Generator",
# "openmoss/Qwen3-8B-ABC",
# "krisfu/medical_r1_qwen3_sft",
# "lxh123666/qwen-0.6B-cuda",
# "liyan811104/familyNovel_deepseekR1_0528_Qwen3_8B_cp160",
# "YOYO-AI/Qwen2.5-14B-YOYO-stock",
# "YOYO-AI/Qwen2.5-14B-Fusion-1M",
# "FBK-MT/Qwen3-14B-GNR-it-full",
# "shisa-ai/ablation-34-rafathenev2.unphi45e6-shisa-v2-unphi-4-14b",
# "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b",
# "nanbeige/Nanbeige4-3B-Base",
# "shisa-ai/ablation-62-dpo.notseq-ablation-60-a55.dpo.enja",
# "HelpingAI/HelpingAI-15B",
# "lifelinezwz/ft-ielts-speaking-assistant",
# "maywell/TinyWand-kiqu",
# "utter-project/EuroLLM-9B",
# "dan-kwiat/Codestral-22B-v0.1-hf-FIM-fix-awq",
# "mesolitica/tinyllama-1.1b-4096-fpf",
# "llm-jp/optimal-sparsity-code-d1024-E32-k2-3.5B-A470M",
# "miromind-ai/MiroThinker-8B-SFT-v0.2",
# "maywell/l3-211m",
# "llm-jp/optimal-sparsity-code-d512-E128-k2-3.3B-A170M",
# "typhoona/sft2_internvl3",
# "v000000/Qwen2.5-14B-Gutenberg-Instruct-Slerpeno",
# "unsloth/GLM-Z1-9B-0414",
# "sh2orc/llama-3-korean-8b-awq",
# "jsincn/phi-3-mini-128k-instruct-awq",
# "utter-project/EuroLLM-9B-Instruct",
# "flux-inc/Flux-Japanese-Qwen2.5-32B-Instruct-V1.0",
# "allegrolab/hubble-8b-500b_toks-standard-hf",
# "laion/r2egym-nl2bashseq",
# "soob3123/amoral-gemma3-12B-v2-qat",
# "jianfeng777/jianfeng",
# "winglian/qwen3-14b-math",
"Neelectric/Llama-3.1-8B-Instruct_SDFT_mathv00.05",
"mremila/Llama-3.1-8B-knowledge",
"llm-jp/optimal-sparsity-math-d2048-E64-k16-26.4B-A7.1B",
"mremila/Llama-3.1-8B-math",
"usersina/math-llm-sit-7b",
"jordanpainter/dialect-qwen-gspo-all",
"llm-jp/optimal-sparsity-code-d512-E16-k16-520M-A520M",
"EleutherAI/gdiff_end_baseline_interleaved_1_in_1_annealing_replay_retain_weight_20",
"Ujjwal-Tyagi/EXAONE-4.0-32B",
"DevQuasar/coma-7B-v0.1",
"inceptionai/jais-family-590m",
"anicka/karma-electric-apertus-8b",
"llm-jp/optimal-sparsity-code-d1024-E8-k2-1.1B-A470M",
"AI-ModelScope/FunReason-MT",
"meditsolutions/MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prune",
"miromind-ai/MiroThinker-14B-DPO-v0.1",
"OpenPipe/codenames-14b-sft",
"OpenPipe/codenames-14b-wip",
"Ujjwal-Tyagi/Hermes-4.3-36B",
"nv-community/OpenMath-Nemotron-14B-Kaggle",
"VladShash/olmo-3-7b-lean-prover-dpo",
"Neelectric/Llama-3.1-8B-Instruct_SDFT_mathv00.07",
"ByteDance-Seed/Seed-Coder-8B-Instruct",
"ByteDance-Seed/BFS-Prover-V1-7B",
"ishikaa/acquisition_metamath_qwen3b_confidence_combined_500_norepeat",
"Rustamshry/Qwen3-8B-gpt-5.4-Reasoning-Distilled",
"Neelectric/Llama-3.1-8B-Instruct_SDFT_mathv00.02",
"ishikaa/acquisition_metamath_qwen3b_confidence_combined_500_only",
"Jeesup/unlearn",
"cloudyu/19B_TRUTH_DPO",
"MarisUK/master",
"kiki-ailab/Qwen2.5-0.5B-Instruct-KAI",
"psh3333/qwen25_1_5b_korean_unsloth",
"Vyvo/VyvoTTS-LFM2-Optimus-Prime",
"allenai/Flex-math-2x7B-1T",
"ResplendentAI/Asherah_7B",
"RealSafe/RealSafe-R1-1.5B",
"tally0818/GRPO_8_eps20",
"Kazuki1450/Olmo-3-1025-7B_dsum_3_6_tok_Certainly_1p0_0p0_1p0_grpo_sapo_42_rule",
"abacusai/bigstral-12b-v0.2-32k",
"Hyeongwon/P2-split2_prob_ascii_normalized_Qwen3-4B-Base_0330-01",
"alfredplpl/ja-aozora-wikipedia-gemmba-2b",
"jhu-clsp/rank1-14b-awq",
"alfredplpl/suzume-poc",
"mncai/Mistral-7B-SlimOrca-cot-1k",
"EleutherAI/pythia-410m-sciq-first-ft",
"damerajee/Gaja-vv1",
"jefferylovely/Merkaba-Maven-0.1",
"JoaoReiz/Llama3.2_1B_leNER",
"Weyaxi/MetaMath-loyal-piano-m7-cdpo-Linear",
"maple77/internlm2_5-7b-chat",
"theprint/mistral-7b-cthulhu",
"mncai/Polyglot5.8B-ShareGPT-Wiki-News_epoch4",
"hishab/titulm-llama-3.2-3b-v2.0-Instruct-v0.1",
"rubraAI/Gemma-1.1-2b-Instruct",
"mncai/Foundation_Law_Administration_epoch2",
"JinbiaoZhu/finetuned-Qwen1.5-0.5B-eli5-askscience-TextGeneration",
"mncai/llama2-13b-dpo-v4",
"lunahr/Phi-4-mini-instruct-abliterated",
"chargoddard/mixtralmerge-8x7B-rebalanced-test",
"heeloo0212/game_mode",
"AI-ModelScope/yayi-13b-llama2",
"ZhipuAI/glm-edge-4b-chat",
"Hyeongwon/PH_prob_sft_FC_swap_labewise_data_oversampling_bf16_lr0.00002_context_12k-Qwen3-8B-Base",
"zhantz/Qwen3-0.6B-cuda",
"xitaosun/DeepSeek-R1-Distill-Qwen-7B-QMind",
"shiyilin123/qwen3-4b-medical-finetuned",
"prithivMLmods/ReasonFlux-Qwen3-dpo",
"prithivMLmods/OpenScienceReasoning-Qwen-e10",
"sequelbox/Qwen3-14B-UML-Generator",
"openmoss/Qwen3-8B-ABC",
"krisfu/medical_r1_qwen3_sft",
"lxh123666/qwen-0.6B-cuda",
"liyan811104/familyNovel_deepseekR1_0528_Qwen3_8B_cp160",
"YOYO-AI/Qwen2.5-14B-YOYO-stock",
"YOYO-AI/Qwen2.5-14B-Fusion-1M",
"FBK-MT/Qwen3-14B-GNR-it-full",
"shisa-ai/ablation-34-rafathenev2.unphi45e6-shisa-v2-unphi-4-14b",
"shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b",
"nanbeige/Nanbeige4-3B-Base",
"shisa-ai/ablation-62-dpo.notseq-ablation-60-a55.dpo.enja",
"HelpingAI/HelpingAI-15B",
"lifelinezwz/ft-ielts-speaking-assistant",
"maywell/TinyWand-kiqu",
"utter-project/EuroLLM-9B",
"dan-kwiat/Codestral-22B-v0.1-hf-FIM-fix-awq",
"mesolitica/tinyllama-1.1b-4096-fpf",
"llm-jp/optimal-sparsity-code-d1024-E32-k2-3.5B-A470M",
"miromind-ai/MiroThinker-8B-SFT-v0.2",
"maywell/l3-211m",
"llm-jp/optimal-sparsity-code-d512-E128-k2-3.3B-A170M",
"typhoona/sft2_internvl3",
"v000000/Qwen2.5-14B-Gutenberg-Instruct-Slerpeno",
"unsloth/GLM-Z1-9B-0414",
"sh2orc/llama-3-korean-8b-awq",
"jsincn/phi-3-mini-128k-instruct-awq",
"utter-project/EuroLLM-9B-Instruct",
"flux-inc/Flux-Japanese-Qwen2.5-32B-Instruct-V1.0",
"allegrolab/hubble-8b-500b_toks-standard-hf",
"laion/r2egym-nl2bashseq",
"soob3123/amoral-gemma3-12B-v2-qat",
"jianfeng777/jianfeng",
"winglian/qwen3-14b-math",
"bralynn/dt.md5.6think1.128.256.mdsteps25.think124",
"prithivMLmods/Geminorum-Wasat-14B-Instruct",
"Tesslate/UIGEN-T2-7B",