chaiml-mistral32-deepse_25601

developer_uid: NischayDnk

submission_id: chaiml-mistral32-deepse_25601_v2

model_name: chaiml-mistral32-deepse_25601_v2

model_group: ChaiML/mistral32_deepsee

status: torndown

timestamp: 2025-07-07T20:12:06+00:00

num_battles: 8331

num_wins: 4432

celo_rating: 1309.9

family_friendly_score: 0.4928

family_friendly_standard_error: 0.007070334645545429

submission_type: basic

model_repo: ChaiML/mistral32_deepseek_cpo_mer_exp2

model_architecture: MistralForCausalLM

model_num_parameters: 24096691200.0

best_of: 6

max_input_tokens: 1536

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.4860981888767486, 'latency_mean': 2.057098183631897, 'latency_p50': 2.0362303256988525, 'latency_p90': 2.282247257232666}, {'batch_size': 3, 'throughput': 0.9480479561903757, 'latency_mean': 3.1567978751659393, 'latency_p50': 3.175247073173523, 'latency_p90': 3.4227890729904176}, {'batch_size': 5, 'throughput': 1.189731618666858, 'latency_mean': 4.177679287195206, 'latency_p50': 4.151451349258423, 'latency_p90': 4.6349050283432005}, {'batch_size': 6, 'throughput': 1.2677028663091725, 'latency_mean': 4.704415488243103, 'latency_p50': 4.6816346645355225, 'latency_p90': 5.239987230300903}, {'batch_size': 8, 'throughput': 1.3766370809994115, 'latency_mean': 5.7648805129528045, 'latency_p50': 5.776118874549866, 'latency_p90': 6.477387714385986}, {'batch_size': 10, 'throughput': 1.4539225524692376, 'latency_mean': 6.814716904163361, 'latency_p50': 6.805126905441284, 'latency_p90': 7.638344407081604}]

gpu_counts: {'NVIDIA A100-SXM4-80GB': 1}

display_name: chaiml-mistral32-deepse_25601_v2

is_internal_developer: False

language_model: ChaiML/mistral32_deepseek_cpo_mer_exp2

model_size: 24B

ranking_group: single

throughput_3p7s: 1.1

us_pacific_date: 2025-07-07

win_ratio: 0.5319889569079342

generation_params: {'temperature': 0.55, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 60, 'presence_penalty': 0.4, 'frequency_penalty': 0.4, 'stopping_words': ['<|im_start|>', '<|im_end|>', '\n'], 'max_input_tokens': 1536, 'best_of': 6, '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': True}

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 chaiml-mistral32-deepse-25601-v2-mkmlizer
Waiting for job on chaiml-mistral32-deepse-25601-v2-mkmlizer to finish
chaiml-mistral32-deepse-25601-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║                                                                     ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║                                                                     ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  Version: 0.29.15                                                   ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║                                                                     ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  belonging to:                                                      ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║                                                                     ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ║                                                                     ║
chaiml-mistral32-deepse-25601-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-mistral32-deepse-25601-v2-mkmlizer: Downloaded to shared memory in 83.052s
chaiml-mistral32-deepse-25601-v2-mkmlizer: Checking if ChaiML/mistral32_deepseek_cpo_mer_exp2 already exists in ChaiML
chaiml-mistral32-deepse-25601-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmps8yky3ql, device:0
chaiml-mistral32-deepse-25601-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-mistral32-deepse-25601-v2-mkmlizer: quantized model in 66.866s
chaiml-mistral32-deepse-25601-v2-mkmlizer: Processed model ChaiML/mistral32_deepseek_cpo_mer_exp2 in 149.918s
chaiml-mistral32-deepse-25601-v2-mkmlizer: creating bucket guanaco-mkml-models
chaiml-mistral32-deepse-25601-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-mistral32-deepse-25601-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia
chaiml-mistral32-deepse-25601-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia/special_tokens_map.json
chaiml-mistral32-deepse-25601-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia/config.json
chaiml-mistral32-deepse-25601-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia/tokenizer_config.json
chaiml-mistral32-deepse-25601-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia/tokenizer.json
chaiml-mistral32-deepse-25601-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia/flywheel_model.1.safetensors
chaiml-mistral32-deepse-25601-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-mistral32-deepse-25601-v2/nvidia/flywheel_model.0.safetensors
chaiml-mistral32-deepse-25601-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/363 [00:09<28:27,  4.73s/it]
Loading 0:   1%|▏         | 5/363 [00:09<09:07,  1.53s/it]
Loading 0:   2%|▏         | 7/363 [00:09<05:41,  1.04it/s]
Loading 0:   4%|▎         | 13/363 [00:09<02:10,  2.68it/s]
Loading 0:   4%|▍         | 16/363 [00:10<01:36,  3.61it/s]
Loading 0:   6%|▌         | 22/363 [00:10<00:59,  5.75it/s]
Loading 0:   7%|▋         | 24/363 [00:10<00:54,  6.27it/s]
Loading 0:   9%|▊         | 31/363 [00:10<00:31, 10.61it/s]
Loading 0:   9%|▉         | 34/363 [00:10<00:29, 11.31it/s]
Loading 0:  10%|█         | 38/363 [00:11<00:22, 14.28it/s]
Loading 0:  11%|█▏        | 41/363 [00:11<00:20, 15.98it/s]
Loading 0:  12%|█▏        | 44/363 [00:11<00:18, 16.98it/s]
Loading 0:  13%|█▎        | 49/363 [00:11<00:15, 20.65it/s]
Loading 0:  14%|█▍        | 52/363 [00:11<00:16, 19.00it/s]
Loading 0:  15%|█▌        | 56/363 [00:11<00:13, 22.43it/s]
Loading 0:  16%|█▋        | 59/363 [00:12<00:18, 16.01it/s]
Loading 0:  17%|█▋        | 62/363 [00:12<00:17, 16.74it/s]
Loading 0:  18%|█▊        | 65/363 [00:12<00:15, 18.79it/s]
Loading 0:  19%|█▊        | 68/363 [00:12<00:14, 20.29it/s]
Loading 0:  20%|█▉        | 71/363 [00:12<00:14, 20.49it/s]
Loading 0:  20%|██        | 74/363 [00:12<00:12, 22.47it/s]
Loading 0:  21%|██        | 77/363 [00:12<00:12, 23.02it/s]
Loading 0:  22%|██▏       | 80/363 [00:13<00:12, 22.22it/s]
Loading 0:  23%|██▎       | 83/363 [00:13<00:11, 23.84it/s]
Loading 0:  24%|██▎       | 86/363 [00:13<00:11, 24.18it/s]
Loading 0:  25%|██▍       | 89/363 [00:13<00:11, 23.00it/s]
Loading 0:  25%|██▌       | 92/363 [00:13<00:11, 23.55it/s]
Loading 0:  26%|██▌       | 95/363 [00:13<00:11, 23.82it/s]
Loading 0:  27%|██▋       | 98/363 [00:14<00:17, 14.87it/s]
Loading 0:  28%|██▊       | 101/363 [00:14<00:14, 17.48it/s]
Loading 0:  29%|██▊       | 104/363 [00:14<00:13, 19.06it/s]
Loading 0:  29%|██▉       | 107/363 [00:14<00:13, 19.69it/s]
Loading 0:  30%|███       | 110/363 [00:14<00:11, 21.58it/s]
Loading 0:  31%|███       | 113/363 [00:14<00:11, 22.50it/s]
Loading 0:  32%|███▏      | 116/363 [00:14<00:11, 21.90it/s]
Loading 0:  33%|███▎      | 119/363 [00:14<00:10, 23.32it/s]
Loading 0:  34%|███▎      | 122/363 [00:15<00:10, 23.71it/s]
Loading 0:  34%|███▍      | 125/363 [00:15<00:10, 22.37it/s]
Loading 0:  35%|███▌      | 128/363 [00:15<00:09, 23.76it/s]
Loading 0:  36%|███▌      | 131/363 [00:15<00:09, 23.65it/s]
Loading 0:  37%|███▋      | 134/363 [00:15<00:10, 22.84it/s]
Loading 0:  38%|███▊      | 137/363 [00:15<00:09, 24.21it/s]
Loading 0:  39%|███▊      | 140/363 [00:15<00:14, 15.91it/s]
Loading 0:  39%|███▉      | 142/363 [00:16<00:14, 15.28it/s]
Loading 0:  40%|████      | 146/363 [00:16<00:10, 19.81it/s]
Loading 0:  41%|████      | 149/363 [00:16<00:10, 20.44it/s]
Loading 0:  42%|████▏     | 152/363 [00:16<00:10, 20.55it/s]
Loading 0:  43%|████▎     | 155/363 [00:16<00:09, 22.67it/s]
Loading 0:  44%|████▎     | 158/363 [00:16<00:08, 23.45it/s]
Loading 0:  44%|████▍     | 161/363 [00:16<00:09, 22.29it/s]
Loading 0:  45%|████▌     | 164/363 [00:16<00:08, 23.90it/s]
Loading 0:  46%|████▌     | 167/363 [00:17<00:08, 24.32it/s]
Loading 0:  47%|████▋     | 170/363 [00:17<00:08, 22.92it/s]
Loading 0:  48%|████▊     | 173/363 [00:17<00:07, 24.36it/s]
Loading 0:  48%|████▊     | 176/363 [00:17<00:12, 15.53it/s]
Loading 0:  49%|████▉     | 179/363 [00:17<00:11, 16.37it/s]
Loading 0:  50%|█████     | 182/363 [00:17<00:09, 18.56it/s]
Loading 0:  51%|█████     | 185/363 [00:18<00:09, 19.73it/s]
Loading 0:  51%|█████     | 185/363 [00:29<00:09, 19.73it/s]
Loading 0:  51%|█████     | 186/363 [00:37<07:25,  2.51s/it]
Loading 0:  53%|█████▎    | 191/363 [00:38<03:50,  1.34s/it]
Loading 0:  53%|█████▎    | 194/363 [00:38<02:44,  1.03it/s]
Loading 0:  54%|█████▍    | 197/363 [00:38<01:56,  1.42it/s]
Loading 0:  56%|█████▌    | 202/363 [00:38<01:08,  2.33it/s]
Loading 0:  56%|█████▋    | 205/363 [00:38<00:52,  3.00it/s]
Loading 0:  58%|█████▊    | 209/363 [00:38<00:35,  4.30it/s]
Loading 0:  58%|█████▊    | 212/363 [00:38<00:27,  5.46it/s]
Loading 0:  59%|█████▉    | 215/363 [00:39<00:25,  5.82it/s]
Loading 0:  60%|██████    | 218/363 [00:39<00:19,  7.45it/s]
Loading 0:  61%|██████    | 221/363 [00:39<00:15,  9.33it/s]
Loading 0:  62%|██████▏   | 224/363 [00:39<00:12, 10.96it/s]
Loading 0:  63%|██████▎   | 227/363 [00:39<00:10, 13.22it/s]
Loading 0:  63%|██████▎   | 230/363 [00:40<00:08, 15.21it/s]
Loading 0:  64%|██████▍   | 233/363 [00:40<00:08, 16.18it/s]
Loading 0:  65%|██████▌   | 236/363 [00:40<00:07, 18.13it/s]
Loading 0:  66%|██████▌   | 239/363 [00:40<00:06, 19.73it/s]
Loading 0:  67%|██████▋   | 242/363 [00:40<00:06, 19.88it/s]
Loading 0:  67%|██████▋   | 245/363 [00:40<00:05, 21.85it/s]
Loading 0:  68%|██████▊   | 248/363 [00:40<00:05, 22.15it/s]
Loading 0:  69%|██████▉   | 251/363 [00:40<00:05, 20.59it/s]
Loading 0:  70%|██████▉   | 254/363 [00:41<00:04, 22.12it/s]
Loading 0:  71%|███████   | 257/363 [00:41<00:07, 14.54it/s]
Loading 0:  71%|███████▏  | 259/363 [00:41<00:07, 14.06it/s]
Loading 0:  72%|███████▏  | 263/363 [00:41<00:05, 18.27it/s]
Loading 0:  73%|███████▎  | 266/363 [00:41<00:04, 19.83it/s]
Loading 0:  74%|███████▍  | 269/363 [00:41<00:04, 20.01it/s]
Loading 0:  75%|███████▍  | 272/363 [00:42<00:04, 21.72it/s]
Loading 0:  76%|███████▌  | 275/363 [00:42<00:03, 22.29it/s]
Loading 0:  77%|███████▋  | 278/363 [00:42<00:03, 21.30it/s]
Loading 0:  77%|███████▋  | 281/363 [00:42<00:03, 23.22it/s]
Loading 0:  78%|███████▊  | 284/363 [00:42<00:03, 22.66it/s]
Loading 0:  79%|███████▉  | 287/363 [00:42<00:03, 21.35it/s]
Loading 0:  80%|███████▉  | 290/363 [00:42<00:03, 23.00it/s]
Loading 0:  81%|████████  | 293/363 [00:43<00:04, 15.16it/s]
Loading 0:  81%|████████▏ | 295/363 [00:43<00:04, 14.75it/s]
Loading 0:  82%|████████▏ | 299/363 [00:43<00:03, 18.73it/s]
Loading 0:  83%|████████▎ | 302/363 [00:43<00:03, 19.40it/s]
Loading 0:  84%|████████▍ | 305/363 [00:43<00:03, 18.71it/s]
Loading 0:  85%|████████▍ | 308/363 [00:43<00:02, 20.45it/s]
Loading 0:  86%|████████▌ | 311/363 [00:44<00:02, 20.95it/s]
Loading 0:  87%|████████▋ | 314/363 [00:44<00:02, 20.34it/s]
Loading 0:  87%|████████▋ | 317/363 [00:44<00:02, 21.68it/s]
Loading 0:  88%|████████▊ | 320/363 [00:44<00:01, 21.83it/s]
Loading 0:  89%|████████▉ | 323/363 [00:44<00:01, 20.08it/s]
Loading 0:  90%|████████▉ | 326/363 [00:44<00:01, 21.33it/s]
Loading 0:  91%|█████████ | 329/363 [00:44<00:01, 21.67it/s]
Loading 0:  91%|█████████▏| 332/363 [00:45<00:02, 13.86it/s]
Loading 0:  92%|█████████▏| 335/363 [00:45<00:01, 16.24it/s]
Loading 0:  93%|█████████▎| 338/363 [00:45<00:01, 17.37it/s]
Loading 0:  94%|█████████▍| 341/363 [00:45<00:01, 17.82it/s]
Loading 0:  95%|█████████▍| 344/363 [00:45<00:00, 19.78it/s]
Loading 0:  96%|█████████▌| 347/363 [00:45<00:00, 20.21it/s]
Loading 0:  96%|█████████▋| 350/363 [00:46<00:00, 19.61it/s]
Loading 0:  97%|█████████▋| 353/363 [00:46<00:00, 20.99it/s]
Loading 0:  98%|█████████▊| 356/363 [00:46<00:00, 21.08it/s]
Loading 0:  99%|█████████▉| 359/363 [00:46<00:00, 20.27it/s]
Loading 0: 100%|█████████▉| 362/363 [00:46<00:00, 21.78it/s]
                                                            
Job chaiml-mistral32-deepse-25601-v2-mkmlizer completed after 187.47s with status: succeeded
Stopping job with name chaiml-mistral32-deepse-25601-v2-mkmlizer
Pipeline stage MKMLizer completed in 188.05s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.19s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-mistral32-deepse-25601-v2
Waiting for inference service chaiml-mistral32-deepse-25601-v2 to be ready
Inference service chaiml-mistral32-deepse-25601-v2 ready after 190.7745921611786s
Pipeline stage MKMLDeployer completed in 191.31s
run pipeline stage %s
Running pipeline stage StressChecker
HTTPConnectionPool(host='guanaco-submitter.guanaco-backend.k2.chaiverse.com', port=80): Read timed out. (read timeout=20)
Received unhealthy response to inference request!
Received healthy response to inference request in 3.335034132003784s
Received healthy response to inference request in 1.8690228462219238s
Received healthy response to inference request in 2.3083982467651367s
Received healthy response to inference request in 2.0140647888183594s
5 requests
1 failed requests
5th percentile: 1.898031234741211
10th percentile: 1.927039623260498
20th percentile: 1.9850564002990723
30th percentile: 2.072931480407715
40th percentile: 2.190664863586426
50th percentile: 2.3083982467651367
60th percentile: 2.7190526008605955
70th percentile: 3.1297069549560543
80th percentile: 6.6982326507568395
90th percentile: 13.424629688262941
95th percentile: 16.787828207015988
99th percentile: 19.47838702201843
mean time: 5.93550934791565
%s, retrying in %s seconds...
Received healthy response to inference request in 2.247659921646118s
Received healthy response to inference request in 1.9977200031280518s
Received healthy response to inference request in 2.483517646789551s
Received healthy response to inference request in 1.813459873199463s
Received healthy response to inference request in 1.9556865692138672s
5 requests
0 failed requests
5th percentile: 1.8419052124023438
10th percentile: 1.8703505516052246
20th percentile: 1.9272412300109862
30th percentile: 1.9640932559967041
40th percentile: 1.9809066295623778
50th percentile: 1.9977200031280518
60th percentile: 2.0976959705352782
70th percentile: 2.1976719379425047
80th percentile: 2.294831466674805
90th percentile: 2.389174556732178
95th percentile: 2.436346101760864
99th percentile: 2.4740833377838136
mean time: 2.09960880279541
Pipeline stage StressChecker completed in 43.11s
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 0.69s
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 0.98s
Shutdown handler de-registered
chaiml-mistral32-deepse_25601_v2 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-mistral32-deepse-25601-v2-profiler
Waiting for inference service chaiml-mistral32-deepse-25601-v2-profiler to be ready
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 6292.61s
Shutdown handler de-registered
chaiml-mistral32-deepse_25601_v2 status is now inactive due to auto deactivation removed underperforming models
chaiml-mistral32-deepse_25601_v2 status is now torndown due to DeploymentManager action