chaiml-gemma3-27b-base-s_9210

developer_uid: NischayDnk

submission_id: chaiml-gemma3-27b-base-s_9210_v4

model_name: chaiml-gemma3-27b-base-s_9210_v4

model_group: ChaiML/gemma3-27b-base-s

status: torndown

timestamp: 2025-04-25T07:20:41+00:00

num_battles: 12428

num_wins: 5140

celo_rating: 1225.89

family_friendly_score: 0.5893999999999999

family_friendly_standard_error: 0.006957120668782452

submission_type: basic

model_repo: ChaiML/gemma3-27b-base-sft-exp3

model_architecture: Gemma3ForCausalLM

model_num_parameters: 29698212096.0

best_of: 1

max_input_tokens: 720

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.31053165742087974, 'latency_mean': 3.2202147996425627, 'latency_p50': 3.2303448915481567, 'latency_p90': 3.5823033332824705}, {'batch_size': 3, 'throughput': 0.7378926923350795, 'latency_mean': 4.064180445671082, 'latency_p50': 4.090712785720825, 'latency_p90': 4.494876456260681}, {'batch_size': 5, 'throughput': 1.0947143538849387, 'latency_mean': 4.52815153837204, 'latency_p50': 4.522315740585327, 'latency_p90': 5.01194429397583}, {'batch_size': 6, 'throughput': 1.2725260415153647, 'latency_mean': 4.674843323230744, 'latency_p50': 4.679685950279236, 'latency_p90': 5.234045457839966}, {'batch_size': 8, 'throughput': 1.5571942424194678, 'latency_mean': 5.094843944311142, 'latency_p50': 5.081257939338684, 'latency_p90': 5.755736756324768}, {'batch_size': 10, 'throughput': 1.7818568040691363, 'latency_mean': 5.562449674606324, 'latency_p50': 5.522664308547974, 'latency_p90': 6.2717695236206055}]

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

display_name: chaiml-gemma3-27b-base-s_9210_v4

is_internal_developer: False

language_model: ChaiML/gemma3-27b-base-sft-exp3

model_size: 30B

ranking_group: single

throughput_3p7s: 0.58

us_pacific_date: 2025-04-25

win_ratio: 0.4135822336659157

generation_params: {'temperature': 0.45, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 40, 'presence_penalty': 0.4, 'frequency_penalty': 0.4, 'stopping_words': ['\n', '<|im_start|>', '<|im_end|>'], 'max_input_tokens': 720, 'best_of': 1, '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 chaiml-gemma3-27b-base-s-9210-v4-mkmlizer
Waiting for job on chaiml-gemma3-27b-base-s-9210-v4-mkmlizer to finish
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║                                                                     ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║                                                                     ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  Version: 0.28.3                                                    ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║                                                                     ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  belonging to:                                                      ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║                                                                     ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ║                                                                     ║
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission chaiml-gemma3-27b-base-s_9210_v3: HTTPConnectionPool(host='chaiml-gemma3-27b-base-s-9210-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: Downloaded to shared memory in 61.472s
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp5hvt00cu, device:0
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission chaiml-gemma3-27b-base-s_9210_v3: HTTPConnectionPool(host='chaiml-gemma3-27b-base-s-9210-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Failed to get response for submission chaiml-gemma3-27b-base-s_9210_v3: HTTPConnectionPool(host='chaiml-gemma3-27b-base-s-9210-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: quantized model in 61.320s
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: Processed model ChaiML/gemma3-27b-base-sft-exp3 in 122.793s
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: creating bucket guanaco-mkml-models
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/config.json
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/added_tokens.json
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/special_tokens_map.json
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/tokenizer_config.json
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/tokenizer.model
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/tokenizer.json
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/flywheel_model.2.safetensors
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/flywheel_model.0.safetensors
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-gemma3-27b-base-s-9210-v4/flywheel_model.1.safetensors
chaiml-gemma3-27b-base-s-9210-v4-mkmlizer: 
Loading 0:   0%|          | 0/808 [00:00<?, ?it/s]
Loading 0:   1%|          | 5/808 [00:00<00:17, 47.03it/s]
Loading 0:   2%|▏         | 14/808 [00:00<00:12, 63.25it/s]
Loading 0:   3%|▎         | 21/808 [00:00<00:17, 44.32it/s]
Loading 0:   3%|▎         | 28/808 [00:00<00:16, 47.94it/s]
Loading 0:   4%|▍         | 35/808 [00:00<00:20, 37.94it/s]
Loading 0:   5%|▌         | 42/808 [00:00<00:17, 44.38it/s]
Loading 0:   7%|▋         | 53/808 [00:01<00:14, 52.50it/s]
Loading 0:   7%|▋         | 59/808 [00:01<00:14, 52.09it/s]
Loading 0:   8%|▊         | 68/808 [00:01<00:13, 55.07it/s]
Loading 0:  10%|▉         | 79/808 [00:01<00:12, 60.47it/s]
Loading 0:  11%|█         | 86/808 [00:01<00:12, 60.10it/s]
Loading 0:  12%|█▏        | 94/808 [00:01<00:11, 60.67it/s]
Loading 0:  13%|█▎        | 105/808 [00:01<00:10, 63.95it/s]
Loading 0:  14%|█▍        | 112/808 [00:02<00:11, 61.40it/s]
Loading 0:  15%|█▍        | 119/808 [00:02<00:15, 43.90it/s]
Loading 0:  15%|█▌        | 125/808 [00:02<00:14, 46.92it/s]
Loading 0:  16%|█▌        | 131/808 [00:02<00:14, 47.74it/s]
Loading 0:  17%|█▋        | 138/808 [00:02<00:12, 52.17it/s]
Loading 0:  18%|█▊        | 144/808 [00:02<00:12, 51.94it/s]
Loading 0:  19%|█▉        | 153/808 [00:02<00:11, 55.83it/s]
Loading 0:  20%|██        | 164/808 [00:03<00:10, 60.02it/s]
Loading 0:  21%|██        | 171/808 [00:03<00:11, 57.86it/s]
Loading 0:  22%|██▏       | 179/808 [00:03<00:10, 60.42it/s]
Loading 0:  24%|██▎       | 190/808 [00:03<00:09, 62.88it/s]
Loading 0:  24%|██▍       | 197/808 [00:03<00:12, 48.29it/s]
Loading 0:  25%|██▌       | 203/808 [00:03<00:12, 50.35it/s]
Loading 0:  26%|██▌       | 211/808 [00:03<00:11, 52.25it/s]
Loading 0:  27%|██▋       | 222/808 [00:04<00:10, 56.96it/s]
Loading 0:  28%|██▊       | 228/808 [00:04<00:10, 55.52it/s]
Loading 0:  29%|██▉       | 237/808 [00:04<00:09, 58.92it/s]
Loading 0:  31%|███       | 248/808 [00:04<00:08, 63.26it/s]
Loading 0:  32%|███▏      | 255/808 [00:04<00:08, 61.94it/s]
Loading 0:  32%|███▏      | 262/808 [00:04<00:08, 60.89it/s]
Loading 0:  33%|███▎      | 269/808 [00:05<00:11, 46.15it/s]
Loading 0:  34%|███▍      | 276/808 [00:05<00:10, 50.09it/s]
Loading 0:  36%|███▌      | 287/808 [00:05<00:09, 56.54it/s]
Loading 0:  36%|███▋      | 293/808 [00:05<00:09, 53.33it/s]
Loading 0:  37%|███▋      | 301/808 [00:20<00:09, 53.33it/s]
Loading 0:  37%|███▋      | 302/808 [00:20<04:38,  1.82it/s]
Loading 0:  39%|███▊      | 313/808 [00:20<02:55,  2.82it/s]
Loading 0:  39%|███▉      | 319/808 [00:20<02:20,  3.47it/s]
Loading 0:  41%|████      | 328/808 [00:20<01:36,  5.00it/s]
Loading 0:  42%|████▏     | 339/808 [00:20<01:02,  7.52it/s]
Loading 0:  43%|████▎     | 346/808 [00:20<00:48,  9.55it/s]
Loading 0:  44%|████▎     | 352/808 [00:21<00:40, 11.17it/s]
Loading 0:  44%|████▍     | 357/808 [00:21<00:33, 13.40it/s]
Loading 0:  45%|████▌     | 367/808 [00:21<00:22, 19.38it/s]
Loading 0:  47%|████▋     | 378/808 [00:21<00:16, 26.30it/s]
Loading 0:  48%|████▊     | 384/808 [00:21<00:14, 29.54it/s]
Loading 0:  49%|████▊     | 393/808 [00:21<00:11, 36.09it/s]
Loading 0:  50%|█████     | 404/808 [00:22<00:09, 44.34it/s]
Loading 0:  51%|█████     | 411/808 [00:22<00:08, 47.30it/s]
Loading 0:  52%|█████▏    | 418/808 [00:22<00:07, 48.81it/s]
Loading 0:  53%|█████▎    | 425/808 [00:22<00:09, 41.08it/s]
Loading 0:  53%|█████▎    | 432/808 [00:22<00:08, 46.20it/s]
Loading 0:  55%|█████▍    | 443/808 [00:22<00:07, 51.31it/s]
Loading 0:  56%|█████▌    | 449/808 [00:22<00:07, 49.66it/s]
Loading 0:  57%|█████▋    | 458/808 [00:23<00:06, 54.72it/s]
Loading 0:  58%|█████▊    | 469/808 [00:23<00:05, 60.44it/s]
Loading 0:  59%|█████▉    | 476/808 [00:23<00:05, 59.74it/s]
Loading 0:  60%|█████▉    | 484/808 [00:23<00:05, 58.93it/s]
Loading 0:  61%|██████▏   | 495/808 [00:23<00:04, 62.90it/s]
Loading 0:  62%|██████▏   | 502/808 [00:23<00:05, 60.65it/s]
Loading 0:  63%|██████▎   | 509/808 [00:23<00:06, 47.45it/s]
Loading 0:  64%|██████▎   | 515/808 [00:24<00:05, 49.45it/s]
Loading 0:  65%|██████▍   | 523/808 [00:24<00:05, 50.93it/s]
Loading 0:  66%|██████▌   | 534/808 [00:24<00:04, 56.46it/s]
Loading 0:  67%|██████▋   | 540/808 [00:24<00:04, 55.02it/s]
Loading 0:  68%|██████▊   | 547/808 [00:24<00:04, 58.45it/s]
Loading 0:  68%|██████▊   | 553/808 [00:24<00:04, 55.98it/s]
Loading 0:  70%|██████▉   | 562/808 [00:24<00:04, 59.09it/s]
Loading 0:  71%|███████   | 573/808 [00:25<00:03, 62.67it/s]
Loading 0:  72%|███████▏  | 580/808 [00:25<00:03, 60.31it/s]
Loading 0:  73%|███████▎  | 587/808 [00:25<00:04, 47.24it/s]
Loading 0:  73%|███████▎  | 593/808 [00:25<00:04, 49.57it/s]
Loading 0:  74%|███████▍  | 601/808 [00:25<00:04, 50.91it/s]
Loading 0:  76%|███████▌  | 612/808 [00:25<00:03, 56.55it/s]
Loading 0:  76%|███████▋  | 618/808 [00:25<00:03, 54.82it/s]
Loading 0:  78%|███████▊  | 627/808 [00:26<00:03, 59.37it/s]
Loading 0:  79%|███████▉  | 638/808 [00:26<00:02, 62.83it/s]
Loading 0:  80%|███████▉  | 645/808 [00:26<00:02, 61.47it/s]
Loading 0:  81%|████████  | 652/808 [00:26<00:02, 59.59it/s]
Loading 0:  82%|████████▏ | 659/808 [00:26<00:03, 45.85it/s]
Loading 0:  82%|████████▏ | 666/808 [00:26<00:02, 49.82it/s]
Loading 0:  84%|████████▍ | 677/808 [00:26<00:02, 55.89it/s]
Loading 0:  85%|████████▍ | 683/808 [00:27<00:02, 53.08it/s]
Loading 0:  85%|████████▌ | 690/808 [00:27<00:02, 56.20it/s]
Loading 0:  86%|████████▌ | 696/808 [00:27<00:02, 54.27it/s]
Loading 0:  87%|████████▋ | 703/808 [00:27<00:01, 58.06it/s]
Loading 0:  88%|████████▊ | 709/808 [00:27<00:01, 55.35it/s]
Loading 0:  89%|████████▊ | 716/808 [00:27<00:01, 58.52it/s]
Loading 0:  89%|████████▊ | 717/808 [00:41<00:01, 58.52it/s]
Loading 0:  89%|████████▉ | 718/808 [00:41<01:12,  1.25it/s]
Loading 0:  90%|█████████ | 729/808 [00:41<00:34,  2.30it/s]
Loading 0:  91%|█████████ | 735/808 [00:42<00:23,  3.05it/s]
Loading 0:  92%|█████████▏| 741/808 [00:42<00:16,  3.97it/s]
Loading 0:  92%|█████████▏| 746/808 [00:42<00:12,  5.14it/s]
Loading 0:  94%|█████████▎| 757/808 [00:42<00:05,  8.70it/s]
Loading 0:  95%|█████████▌| 768/808 [00:42<00:03, 13.19it/s]
Loading 0:  96%|█████████▌| 775/808 [00:42<00:01, 16.52it/s]
Loading 0:  97%|█████████▋| 783/808 [00:43<00:01, 21.06it/s]
Loading 0:  98%|█████████▊| 794/808 [00:43<00:00, 28.40it/s]
Loading 0:  99%|█████████▉| 801/808 [00:43<00:00, 32.65it/s]
Loading 0: 100%|██████████| 808/808 [00:43<00:00, 37.44it/s]
                                                            
Failed to get response for submission chaiml-gemma3-27b-base-s_9210_v3: HTTPConnectionPool(host='chaiml-gemma3-27b-base-s-9210-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Job chaiml-gemma3-27b-base-s-9210-v4-mkmlizer completed after 157.07s with status: succeeded
Stopping job with name chaiml-gemma3-27b-base-s-9210-v4-mkmlizer
Pipeline stage MKMLizer completed in 157.69s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.23s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-gemma3-27b-base-s-9210-v4
Waiting for inference service chaiml-gemma3-27b-base-s-9210-v4 to be ready
Failed to get response for submission chaiml-gemma3-27b-base-s_9210_v3: HTTPConnectionPool(host='chaiml-gemma3-27b-base-s-9210-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Inference service chaiml-gemma3-27b-base-s-9210-v4 ready after 130.4423213005066s
Pipeline stage MKMLDeployer completed in 131.09s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.8609859943389893s
Received healthy response to inference request in 3.307046413421631s
Received healthy response to inference request in 3.329287528991699s
Received healthy response to inference request in 3.2033023834228516s
Received healthy response to inference request in 3.2863729000091553s
5 requests
0 failed requests
5th percentile: 3.219916486740112
10th percentile: 3.236530590057373
20th percentile: 3.2697587966918946
30th percentile: 3.29050760269165
40th percentile: 3.2987770080566405
50th percentile: 3.307046413421631
60th percentile: 3.3159428596496583
70th percentile: 3.3248393058776857
80th percentile: 3.4356272220611572
90th percentile: 3.6483066082000732
95th percentile: 3.7546463012695312
99th percentile: 3.839718055725098
mean time: 3.3973990440368653
Pipeline stage StressChecker completed in 19.13s
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 1.16s
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 1.00s
Shutdown handler de-registered
chaiml-gemma3-27b-base-s_9210_v4 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 0.14s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-gemma3-27b-base-s-9210-v4-profiler
Waiting for inference service chaiml-gemma3-27b-base-s-9210-v4-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
Pipeline stage OfflineFamilyFriendlyScorer completed in 3403.63s
Shutdown handler de-registered
chaiml-gemma3-27b-base-s_9210_v4 status is now inactive due to auto deactivation removed underperforming models
chaiml-gemma3-27b-base-s_9210_v4 status is now torndown due to DeploymentManager action