qwen-qwen2-5-7b-instruct

developer_uid: bogoconic1

submission_id: qwen-qwen2-5-7b-instruct_v1

model_name: qwen-qwen2-5-7b-instruct_v1

model_group: Qwen/Qwen2.5-7B-Instruct

status: torndown

timestamp: 2025-06-27T06:09:35+00:00

num_battles: 9375

num_wins: 3452

celo_rating: 1185.73

family_friendly_score: 0.7886

family_friendly_standard_error: 0.00577425389119668

submission_type: basic

model_repo: Qwen/Qwen2.5-7B-Instruct

model_architecture: Qwen2ForCausalLM

model_num_parameters: 7615487488.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8872435340877597, 'latency_mean': 1.1268979787826539, 'latency_p50': 1.1218668222427368, 'latency_p90': 1.2483640909194946}, {'batch_size': 4, 'throughput': 1.8427602632341182, 'latency_mean': 2.165526317358017, 'latency_p50': 2.171611785888672, 'latency_p90': 2.3935977935791013}, {'batch_size': 5, 'throughput': 1.9672156367017748, 'latency_mean': 2.532311336994171, 'latency_p50': 2.543743371963501, 'latency_p90': 2.8306395769119264}, {'batch_size': 8, 'throughput': 2.187491938967161, 'latency_mean': 3.625134307146072, 'latency_p50': 3.630725622177124, 'latency_p90': 4.1400621891021725}, {'batch_size': 10, 'throughput': 2.2458677896648163, 'latency_mean': 4.414090141057968, 'latency_p50': 4.439616084098816, 'latency_p90': 4.935709881782532}, {'batch_size': 12, 'throughput': 2.2874107954337615, 'latency_mean': 5.193327207565307, 'latency_p50': 5.22837221622467, 'latency_p90': 5.811381554603576}, {'batch_size': 15, 'throughput': 2.2983859776176776, 'latency_mean': 6.4465145742893215, 'latency_p50': 6.443965435028076, 'latency_p90': 7.260407090187073}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: qwen-qwen2-5-7b-instruct_v1

is_internal_developer: True

language_model: Qwen/Qwen2.5-7B-Instruct

model_size: 8B

ranking_group: single

throughput_3p7s: 2.21

us_pacific_date: 2025-06-26

win_ratio: 0.36821333333333334

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': ['\n', '<|im_end|>', '<|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-qwen2-5-7b-instruct-v1-mkmlizer
Waiting for job on qwen-qwen2-5-7b-instruct-v1-mkmlizer to finish
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║                                                                     ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║                                                                     ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  Version: 0.29.3                                                    ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  https://mk1.ai                                                     ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║                                                                     ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  belonging to:                                                      ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║                                                                     ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  Chai Research Corp.                                                ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ║                                                                     ║
qwen-qwen2-5-7b-instruct-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
qwen-qwen2-5-7b-instruct-v1-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-qwen2-5-7b-instruct-v1-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-qwen2-5-7b-instruct-v1-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-qwen2-5-7b-instruct-v1-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-qwen2-5-7b-instruct-v1-mkmlizer: Downloaded to shared memory in 22.380s
qwen-qwen2-5-7b-instruct-v1-mkmlizer: Checking if Qwen/Qwen2.5-7B-Instruct already exists in ChaiML
qwen-qwen2-5-7b-instruct-v1-mkmlizer: Creating repo ChaiML/Qwen2.5-7B-Instruct and uploading /tmp/tmpdnpzfd43 to it
qwen-qwen2-5-7b-instruct-v1-mkmlizer: 
  0%|          | 0/4 [00:00<?, ?it/s]
 25%|██▌       | 1/4 [00:02<00:08,  2.92s/it]
 50%|█████     | 2/4 [00:06<00:06,  3.39s/it]
 75%|███████▌  | 3/4 [00:12<00:04,  4.35s/it]
100%|██████████| 4/4 [00:15<00:00,  3.77s/it]
100%|██████████| 4/4 [00:15<00:00,  3.75s/it]
qwen-qwen2-5-7b-instruct-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpdnpzfd43, device:0
qwen-qwen2-5-7b-instruct-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
qwen-qwen2-5-7b-instruct-v1-mkmlizer: quantized model in 22.079s
qwen-qwen2-5-7b-instruct-v1-mkmlizer: Processed model Qwen/Qwen2.5-7B-Instruct in 76.963s
qwen-qwen2-5-7b-instruct-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
qwen-qwen2-5-7b-instruct-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/config.json
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/added_tokens.json
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/special_tokens_map.json
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/tokenizer_config.json
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/merges.txt s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/merges.txt
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/vocab.json s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/vocab.json
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/tokenizer.json
qwen-qwen2-5-7b-instruct-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/qwen-qwen2-5-7b-instruct-v1/flywheel_model.0.safetensors
qwen-qwen2-5-7b-instruct-v1-mkmlizer: 
Loading 0:   0%|          | 0/339 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/339 [00:00<00:10, 32.79it/s]
Loading 0:   4%|▍         | 13/339 [00:00<00:06, 46.92it/s]
Loading 0:   5%|▌         | 18/339 [00:00<00:07, 44.66it/s]
Loading 0:   7%|▋         | 25/339 [00:00<00:06, 49.72it/s]
Loading 0:   9%|▉         | 31/339 [00:00<00:06, 50.19it/s]
Loading 0:  11%|█         | 37/339 [00:00<00:06, 50.00it/s]
Loading 0:  13%|█▎        | 43/339 [00:00<00:05, 49.98it/s]
Loading 0:  14%|█▍        | 49/339 [00:01<00:06, 47.01it/s]
Loading 0:  16%|█▌        | 54/339 [00:01<00:06, 44.78it/s]
Loading 0:  18%|█▊        | 61/339 [00:01<00:05, 47.58it/s]
Loading 0:  19%|█▉        | 66/339 [00:01<00:05, 45.89it/s]
Loading 0:  22%|██▏       | 73/339 [00:01<00:05, 48.95it/s]
Loading 0:  24%|██▍       | 82/339 [00:01<00:04, 56.95it/s]
Loading 0:  26%|██▌       | 88/339 [00:01<00:06, 38.85it/s]
Loading 0:  28%|██▊       | 94/339 [00:02<00:05, 41.15it/s]
Loading 0:  29%|██▉       | 99/339 [00:02<00:05, 41.05it/s]
Loading 0:  31%|███▏      | 106/339 [00:02<00:05, 45.49it/s]
Loading 0:  33%|███▎      | 111/339 [00:02<00:05, 44.27it/s]
Loading 0:  35%|███▍      | 118/339 [00:02<00:04, 47.77it/s]
Loading 0:  36%|███▋      | 123/339 [00:02<00:04, 45.06it/s]
Loading 0:  38%|███▊      | 130/339 [00:02<00:04, 44.98it/s]
Loading 0:  41%|████      | 139/339 [00:02<00:03, 55.11it/s]
Loading 0:  43%|████▎     | 145/339 [00:03<00:04, 43.09it/s]
Loading 0:  44%|████▍     | 150/339 [00:03<00:04, 44.16it/s]
Loading 0:  46%|████▌     | 155/339 [00:03<00:04, 42.88it/s]
Loading 0:  47%|████▋     | 160/339 [00:03<00:04, 40.22it/s]
Loading 0:  49%|████▉     | 167/339 [00:03<00:03, 44.65it/s]
Loading 0:  51%|█████     | 172/339 [00:03<00:03, 43.71it/s]
Loading 0:  53%|█████▎    | 179/339 [00:03<00:03, 46.88it/s]
Loading 0:  54%|█████▍    | 184/339 [00:04<00:04, 31.62it/s]
Loading 0:  56%|█████▋    | 191/339 [00:04<00:03, 38.74it/s]
Loading 0:  58%|█████▊    | 196/339 [00:04<00:03, 36.34it/s]
Loading 0:  60%|██████    | 204/339 [00:04<00:03, 40.65it/s]
Loading 0:  62%|██████▏   | 209/339 [00:04<00:03, 35.37it/s]
Loading 0:  64%|██████▍   | 217/339 [00:04<00:02, 42.29it/s]
Loading 0:  65%|██████▌   | 222/339 [00:05<00:02, 42.41it/s]
Loading 0:  68%|██████▊   | 229/339 [00:05<00:02, 45.94it/s]
Loading 0:  69%|██████▉   | 234/339 [00:05<00:02, 44.38it/s]
Loading 0:  71%|███████   | 241/339 [00:05<00:02, 48.18it/s]
Loading 0:  73%|███████▎  | 246/339 [00:05<00:02, 46.28it/s]
Loading 0:  75%|███████▍  | 253/339 [00:05<00:01, 48.88it/s]
Loading 0:  76%|███████▌  | 258/339 [00:05<00:01, 45.78it/s]
Loading 0:  78%|███████▊  | 265/339 [00:05<00:01, 48.07it/s]
Loading 0:  80%|███████▉  | 270/339 [00:06<00:01, 47.97it/s]
Loading 0:  81%|████████▏ | 276/339 [00:06<00:01, 45.82it/s]
Loading 0:  83%|████████▎ | 281/339 [00:06<00:02, 28.64it/s]
Loading 0:  84%|████████▍ | 286/339 [00:06<00:01, 32.21it/s]
Loading 0:  86%|████████▌ | 290/339 [00:06<00:01, 32.45it/s]
Loading 0:  87%|████████▋ | 294/339 [00:06<00:01, 33.11it/s]
Loading 0:  89%|████████▉ | 302/339 [00:07<00:00, 41.26it/s]
Loading 0:  91%|█████████ | 307/339 [00:07<00:00, 41.38it/s]
Loading 0:  93%|█████████▎| 314/339 [00:07<00:00, 46.27it/s]
Loading 0:  94%|█████████▍| 319/339 [00:07<00:00, 45.23it/s]
Loading 0:  96%|█████████▌| 326/339 [00:07<00:00, 48.96it/s]
Loading 0:  98%|█████████▊| 332/339 [00:07<00:00, 49.72it/s]
Loading 0: 100%|█████████▉| 338/339 [00:07<00:00, 49.13it/s]
                                                            
Job qwen-qwen2-5-7b-instruct-v1-mkmlizer completed after 95.82s with status: succeeded
Stopping job with name qwen-qwen2-5-7b-instruct-v1-mkmlizer
Pipeline stage MKMLizer completed in 96.45s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.17s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service qwen-qwen2-5-7b-instruct-v1
Waiting for inference service qwen-qwen2-5-7b-instruct-v1 to be ready
Inference service qwen-qwen2-5-7b-instruct-v1 ready after 150.61687779426575s
Pipeline stage MKMLDeployer completed in 151.20s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8470754623413086s
Received healthy response to inference request in 1.2089941501617432s
Received healthy response to inference request in 1.2152924537658691s
Received healthy response to inference request in 1.1421146392822266s
Received healthy response to inference request in 1.088061809539795s
5 requests
0 failed requests
5th percentile: 1.0988723754882812
10th percentile: 1.1096829414367675
20th percentile: 1.1313040733337403
30th percentile: 1.1554905414581298
40th percentile: 1.1822423458099365
50th percentile: 1.2089941501617432
60th percentile: 1.2115134716033935
70th percentile: 1.2140327930450439
80th percentile: 1.341649055480957
90th percentile: 1.5943622589111328
95th percentile: 1.7207188606262207
99th percentile: 1.821804141998291
mean time: 1.3003077030181884
Pipeline stage StressChecker completed in 8.21s
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 0.72s
Shutdown handler de-registered
qwen-qwen2-5-7b-instruct_v1 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
Pipeline stage OfflineFamilyFriendlyScorer completed in 2489.63s
Shutdown handler de-registered
qwen-qwen2-5-7b-instruct_v1 status is now inactive due to auto deactivation removed underperforming models
qwen-qwen2-5-7b-instruct_v1 status is now torndown due to DeploymentManager action