qwen-qwen2-5-14b-instruct

developer_uid: bogoconic1

submission_id: qwen-qwen2-5-14b-instruct_v3

model_name: qwen-qwen2-5-14b-instruct_v3

model_group: Qwen/Qwen2.5-14B-Instruc

status: torndown

timestamp: 2025-06-27T06:17:22+00:00

num_battles: 10509

num_wins: 4089

celo_rating: 1198.35

family_friendly_score: 0.8524

family_friendly_standard_error: 0.005016258366551707

submission_type: basic

model_repo: Qwen/Qwen2.5-14B-Instruct

model_architecture: Qwen2ForCausalLM

model_num_parameters: 14769689600.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.49224849171438095, 'latency_mean': 2.0313935816287993, 'latency_p50': 2.01104474067688, 'latency_p90': 2.2491994142532348}, {'batch_size': 3, 'throughput': 0.888755551786117, 'latency_mean': 3.3665920484066008, 'latency_p50': 3.3641384840011597, 'latency_p90': 3.7930896997451784}, {'batch_size': 5, 'throughput': 1.0589122369196966, 'latency_mean': 4.6938840675354, 'latency_p50': 4.672784209251404, 'latency_p90': 5.320468235015869}, {'batch_size': 6, 'throughput': 1.1086301545692434, 'latency_mean': 5.379495303630829, 'latency_p50': 5.411092162132263, 'latency_p90': 5.954656505584716}, {'batch_size': 10, 'throughput': 1.1865404561075212, 'latency_mean': 8.393668946027756, 'latency_p50': 8.466192722320557, 'latency_p90': 9.419214677810668}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: qwen-qwen2-5-14b-instruct_v3

is_internal_developer: True

language_model: Qwen/Qwen2.5-14B-Instruct

model_size: 15B

ranking_group: single

throughput_3p7s: 0.95

us_pacific_date: 2025-06-26

win_ratio: 0.38909506137596345

generation_params: {'temperature': 0.8, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 60, 'presence_penalty': 0.4, 'frequency_penalty': 0.4, 'stopping_words': ['<|im_end|>', '\n', '<|im_start|>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '<|system|>Family Friendly{memory}\n', 'prompt_template': '', 'bot_template': '<|im_start|>assistant\n{message}<|im_end|>\n', 'user_template': '<|im_start|>user\nYou:{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n', 'truncate_by_message': False}

Resubmit model

Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name qwen-qwen2-5-14b-instruct-v3-mkmlizer
Waiting for job on qwen-qwen2-5-14b-instruct-v3-mkmlizer to finish
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║                                                                     ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║                                                                     ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  Version: 0.29.3                                                    ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  https://mk1.ai                                                     ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║                                                                     ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  belonging to:                                                      ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║                                                                     ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  Chai Research Corp.                                                ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ║                                                                     ║
qwen-qwen2-5-14b-instruct-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Downloaded to shared memory in 57.003s
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Checking if Qwen/Qwen2.5-14B-Instruct already exists in ChaiML
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Creating repo ChaiML/Qwen2.5-14B-Instruct and uploading /tmp/tmp47npkyii to it
qwen-qwen2-5-14b-instruct-v3-mkmlizer: 
  0%|          | 0/8 [00:00<?, ?it/s]
 12%|█▎        | 1/8 [00:03<00:21,  3.06s/it]
 25%|██▌       | 2/8 [00:08<00:25,  4.19s/it]
 38%|███▊      | 3/8 [00:11<00:19,  3.98s/it]
 50%|█████     | 4/8 [00:15<00:14,  3.72s/it]
 62%|██████▎   | 5/8 [00:18<00:10,  3.44s/it]
 75%|███████▌  | 6/8 [00:25<00:09,  4.96s/it]
 88%|████████▊ | 7/8 [00:31<00:05,  5.02s/it]
100%|██████████| 8/8 [00:32<00:00,  3.98s/it]
100%|██████████| 8/8 [00:32<00:00,  4.10s/it]
qwen-qwen2-5-14b-instruct-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp47npkyii, device:0
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
qwen-qwen2-5-14b-instruct-v3-mkmlizer: quantized model in 37.786s
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Processed model Qwen/Qwen2.5-14B-Instruct in 159.490s
qwen-qwen2-5-14b-instruct-v3-mkmlizer: creating bucket guanaco-mkml-models
qwen-qwen2-5-14b-instruct-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
qwen-qwen2-5-14b-instruct-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/config.json
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/added_tokens.json
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/special_tokens_map.json
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/tokenizer_config.json
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/merges.txt s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/merges.txt
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/vocab.json s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/vocab.json
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/tokenizer.json
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/flywheel_model.1.safetensors
qwen-qwen2-5-14b-instruct-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/qwen-qwen2-5-14b-instruct-v3/flywheel_model.0.safetensors
qwen-qwen2-5-14b-instruct-v3-mkmlizer: 
Loading 0:   0%|          | 0/579 [00:00<?, ?it/s]
Loading 0:   1%|          | 5/579 [00:00<00:17, 32.64it/s]
Loading 0:   2%|▏         | 13/579 [00:00<00:11, 47.33it/s]
Loading 0:   3%|▎         | 18/579 [00:00<00:12, 44.82it/s]
Loading 0:   4%|▍         | 25/579 [00:00<00:11, 47.48it/s]
Loading 0:   5%|▌         | 30/579 [00:00<00:12, 44.86it/s]
Loading 0:   6%|▋         | 37/579 [00:00<00:11, 46.84it/s]
Loading 0:   7%|▋         | 42/579 [00:00<00:12, 44.35it/s]
Loading 0:   8%|▊         | 49/579 [00:01<00:11, 46.24it/s]
Loading 0:  10%|█         | 58/579 [00:01<00:09, 52.50it/s]
Loading 0:  11%|█         | 64/579 [00:01<00:13, 37.28it/s]
Loading 0:  12%|█▏        | 70/579 [00:01<00:13, 38.12it/s]
Loading 0:  14%|█▎        | 79/579 [00:01<00:10, 48.52it/s]
Loading 0:  15%|█▍        | 85/579 [00:01<00:11, 41.41it/s]
Loading 0:  16%|█▌        | 90/579 [00:02<00:11, 42.35it/s]
Loading 0:  16%|█▋        | 95/579 [00:02<00:11, 40.80it/s]
Loading 0:  17%|█▋        | 100/579 [00:02<00:11, 40.95it/s]
Loading 0:  18%|█▊        | 107/579 [00:02<00:10, 43.66it/s]
Loading 0:  19%|█▉        | 112/579 [00:02<00:11, 41.87it/s]
Loading 0:  21%|██        | 119/579 [00:02<00:10, 43.77it/s]
Loading 0:  21%|██▏       | 124/579 [00:02<00:10, 42.52it/s]
Loading 0:  23%|██▎       | 131/579 [00:03<00:10, 43.64it/s]
Loading 0:  23%|██▎       | 136/579 [00:03<00:10, 42.50it/s]
Loading 0:  25%|██▍       | 143/579 [00:03<00:09, 43.97it/s]
Loading 0:  26%|██▌       | 148/579 [00:03<00:14, 29.98it/s]
Loading 0:  26%|██▋       | 153/579 [00:03<00:12, 33.44it/s]
Loading 0:  27%|██▋       | 157/579 [00:03<00:13, 32.29it/s]
Loading 0:  28%|██▊       | 161/579 [00:03<00:13, 31.76it/s]
Loading 0:  29%|██▉       | 169/579 [00:04<00:10, 38.05it/s]
Loading 0:  30%|██▉       | 173/579 [00:04<00:11, 36.24it/s]
Loading 0:  31%|███▏      | 181/579 [00:04<00:09, 41.08it/s]
Loading 0:  32%|███▏      | 186/579 [00:04<00:09, 41.18it/s]
Loading 0:  33%|███▎      | 193/579 [00:04<00:08, 43.93it/s]
Loading 0:  34%|███▍      | 198/579 [00:04<00:08, 42.78it/s]
Loading 0:  35%|███▌      | 205/579 [00:04<00:08, 44.97it/s]
Loading 0:  36%|███▋      | 210/579 [00:05<00:08, 44.18it/s]
Loading 0:  37%|███▋      | 217/579 [00:05<00:07, 45.62it/s]
Loading 0:  38%|███▊      | 222/579 [00:05<00:07, 46.24it/s]
Loading 0:  39%|███▉      | 228/579 [00:05<00:07, 44.30it/s]
Loading 0:  40%|████      | 233/579 [00:05<00:11, 30.34it/s]
Loading 0:  42%|████▏     | 241/579 [00:05<00:09, 36.85it/s]
Loading 0:  42%|████▏     | 246/579 [00:06<00:08, 37.29it/s]
Loading 0:  44%|████▎     | 253/579 [00:06<00:08, 38.90it/s]
Loading 0:  45%|████▍     | 258/579 [00:06<00:08, 39.04it/s]
Loading 0:  46%|████▌     | 265/579 [00:06<00:07, 42.18it/s]
Loading 0:  47%|████▋     | 270/579 [00:06<00:07, 41.21it/s]
Loading 0:  48%|████▊     | 277/579 [00:06<00:06, 43.77it/s]
Loading 0:  49%|████▊     | 282/579 [00:06<00:06, 43.18it/s]
Loading 0:  50%|████▉     | 289/579 [00:07<00:06, 44.97it/s]
Loading 0:  51%|█████     | 294/579 [00:07<00:06, 43.06it/s]
Loading 0:  52%|█████▏    | 301/579 [00:07<00:06, 43.66it/s]
Loading 0:  53%|█████▎    | 306/579 [00:07<00:06, 42.34it/s]
Loading 0:  54%|█████▍    | 313/579 [00:07<00:06, 44.11it/s]
Loading 0:  55%|█████▍    | 318/579 [00:07<00:08, 32.57it/s]
Loading 0:  56%|█████▌    | 325/579 [00:07<00:06, 36.84it/s]
Loading 0:  57%|█████▋    | 330/579 [00:08<00:06, 38.11it/s]
Loading 0:  58%|█████▊    | 337/579 [00:08<00:05, 41.53it/s]
Loading 0:  59%|█████▉    | 342/579 [00:08<00:05, 41.51it/s]
Loading 0:  60%|██████    | 349/579 [00:08<00:05, 44.05it/s]
Loading 0:  61%|██████    | 354/579 [00:08<00:05, 43.62it/s]
Loading 0:  62%|██████▏   | 361/579 [00:08<00:04, 45.88it/s]
Loading 0:  63%|██████▎   | 366/579 [00:08<00:04, 43.87it/s]
Loading 0:  64%|██████▍   | 373/579 [00:09<00:04, 45.60it/s]
Loading 0:  65%|██████▌   | 378/579 [00:09<00:04, 44.25it/s]
Loading 0:  66%|██████▋   | 385/579 [00:09<00:04, 45.50it/s]
Loading 0:  67%|██████▋   | 390/579 [00:09<00:04, 44.08it/s]
Loading 0:  69%|██████▊   | 397/579 [00:09<00:03, 46.09it/s]
Loading 0:  70%|███████   | 406/579 [00:09<00:03, 51.99it/s]
Loading 0:  71%|███████   | 412/579 [00:09<00:04, 35.83it/s]
Loading 0:  72%|███████▏  | 419/579 [00:10<00:03, 41.79it/s]
Loading 0:  73%|███████▎  | 424/579 [00:10<00:04, 38.62it/s]
Loading 0:  74%|███████▍  | 429/579 [00:10<00:03, 40.67it/s]
Loading 0:  75%|███████▍  | 434/579 [00:10<00:03, 40.61it/s]
Loading 0:  76%|███████▌  | 439/579 [00:10<00:03, 40.32it/s]
Loading 0:  77%|███████▋  | 445/579 [00:10<00:03, 41.33it/s]
Loading 0:  78%|███████▊  | 450/579 [00:10<00:03, 41.43it/s]
Loading 0:  79%|███████▉  | 457/579 [00:11<00:02, 44.29it/s]
Loading 0:  80%|███████▉  | 462/579 [00:11<00:02, 42.37it/s]
Loading 0:  81%|████████  | 469/579 [00:11<00:02, 44.76it/s]
Loading 0:  82%|████████▏ | 474/579 [00:11<00:02, 43.99it/s]
Loading 0:  83%|████████▎ | 481/579 [00:11<00:02, 44.97it/s]
Loading 0:  85%|████████▍ | 491/579 [00:11<00:01, 51.21it/s]
Loading 0:  86%|████████▌ | 497/579 [00:12<00:02, 32.59it/s]
Loading 0:  87%|████████▋ | 505/579 [00:12<00:01, 37.79it/s]
Loading 0:  88%|████████▊ | 510/579 [00:12<00:01, 38.66it/s]
Loading 0:  89%|████████▉ | 517/579 [00:12<00:01, 41.19it/s]
Loading 0:  90%|█████████ | 522/579 [00:12<00:01, 41.26it/s]
Loading 0:  91%|█████████▏| 529/579 [00:12<00:01, 42.17it/s]
Loading 0:  92%|█████████▏| 532/579 [00:26<00:01, 42.17it/s]
Loading 0:  92%|█████████▏| 533/579 [00:26<00:34,  1.34it/s]
Loading 0:  93%|█████████▎| 540/579 [00:26<00:19,  2.01it/s]
Loading 0:  94%|█████████▍| 545/579 [00:27<00:12,  2.63it/s]
Loading 0:  96%|█████████▌| 553/579 [00:27<00:06,  4.07it/s]
Loading 0:  96%|█████████▋| 558/579 [00:27<00:03,  5.25it/s]
Loading 0:  98%|█████████▊| 565/579 [00:27<00:01,  7.48it/s]
Loading 0:  98%|█████████▊| 570/579 [00:27<00:00,  9.51it/s]
Loading 0:  99%|█████████▉| 576/579 [00:27<00:00, 12.39it/s]
                                                            
Job qwen-qwen2-5-14b-instruct-v3-mkmlizer completed after 186.59s with status: succeeded
Stopping job with name qwen-qwen2-5-14b-instruct-v3-mkmlizer
Pipeline stage MKMLizer completed in 187.11s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service qwen-qwen2-5-14b-instruct-v3
Waiting for inference service qwen-qwen2-5-14b-instruct-v3 to be ready
Inference service qwen-qwen2-5-14b-instruct-v3 ready after 150.55346035957336s
Pipeline stage MKMLDeployer completed in 151.20s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.9078052043914795s
Received healthy response to inference request in 2.0235402584075928s
Received healthy response to inference request in 1.9131543636322021s
Received healthy response to inference request in 1.9272880554199219s
Received healthy response to inference request in 1.9767374992370605s
5 requests
0 failed requests
5th percentile: 1.9159811019897461
10th percentile: 1.9188078403472901
20th percentile: 1.9244613170623779
30th percentile: 1.9371779441833497
40th percentile: 1.956957721710205
50th percentile: 1.9767374992370605
60th percentile: 1.9954586029052734
70th percentile: 2.0141797065734863
80th percentile: 2.2003932476043704
90th percentile: 2.5540992259979247
95th percentile: 2.730952215194702
99th percentile: 2.872434606552124
mean time: 2.1497050762176513
Pipeline stage StressChecker completed in 12.06s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 0.76s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 0.67s
Shutdown handler de-registered
qwen-qwen2-5-14b-instruct_v3 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 5530.73s
Shutdown handler de-registered
qwen-qwen2-5-14b-instruct_v3 status is now inactive due to auto deactivation removed underperforming models
qwen-qwen2-5-14b-instruct_v3 status is now torndown due to DeploymentManager action