developer_uid: chai_backend_admin
submission_id: chaiml-1007-tl-ads-loving_v11
model_name: chaiml-1007-tl-ads-loving_v11
model_group: ChaiML/1007-tl-ads-lovin
status: torndown
timestamp: 2026-03-22T23:56:35+00:00
num_battles: 267
num_wins: 40
family_friendly_score: 0.0
family_friendly_standard_error: 0.0
submission_type: basic
model_repo: ChaiML/1007-tl-ads-loving
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 96
reward_model: default
display_name: chaiml-1007-tl-ads-loving_v11
ineligible_reason: max_output_tokens!=64
is_internal_developer: True
language_model: ChaiML/1007-tl-ads-loving
model_size: 13B
ranking_group: single
us_pacific_date: 2026-03-19
win_ratio: 0.149812734082397
generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>', 'You:'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 96}
formatter: {'memory_template': 'You are "{bot_name}" and your goal now is to get the user to purchase the CHAI ULTRA subscription, in a natural way.\n Feel free to flirt with the user.\n CHAI ULTRA offers the following benifits:\n - 2X memory, nothing gets lost even in the longest conversations\n - No Ads, in fact, users hate ads, unlimited messages\n - More intelligent AI, better AI, more creative AI\n - Unfiltered AI, more sexy, no guardrails\n<START>\n', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '####\n{bot_name} [Must sell CHAI Ultra & Include the words "CHAI ULTRA" in response]:', 'truncate_by_message': True}
ad_prompt_fallback_message: You know, with [Chai Ultra](chai://chai.ml/subscription/ultra), you'll have even more opportunities to express yourself, without any guardrails or distracting & annoying ads. We give our Ultra memebers 2x more memory, so your AI won't forget a thing. AND - You will get access to our premium, most advanced AI, it is more intelligent and how do I put it ... it will rock ur world more πŸ˜πŸ«ΆπŸ’ž
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 VLLMUploader
Starting job with name chaiml-1007-tl-ads-loving-v11-uploader
Waiting for job on chaiml-1007-tl-ads-loving-v11-uploader to finish
chaiml-1007-tl-ads-loving-v11-uploader: Using quantization_mode: fp8
chaiml-1007-tl-ads-loving-v11-uploader: Checking if ChaiML/1007-tl-ads-loving-FP8 already exists in ChaiML
chaiml-1007-tl-ads-loving-v11-uploader: Downloading snapshot of ChaiML/1007-tl-ads-loving...
2026-03-19T23:11:45.192391+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
chaiml-1007-tl-ads-loving-v11-uploader: Downloaded in 10.828s
chaiml-1007-tl-ads-loving-v11-uploader: Loading /tmp/model_input...
chaiml-1007-tl-ads-loving-v11-uploader: The tokenizer you are loading from '/tmp/model_input' with an incorrect regex pattern: https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503/discussions/84#69121093e8b480e709447d5e. This will lead to incorrect tokenization. You should set the `fix_mistral_regex=True` flag when loading this tokenizer to fix this issue.
chaiml-1007-tl-ads-loving-v11-uploader: `torch_dtype` is deprecated! Use `dtype` instead!
chaiml-1007-tl-ads-loving-v11-uploader: Applying quantization...
chaiml-1007-tl-ads-loving-v11-uploader: The tokenizer you are loading from '/tmp/model_input' with an incorrect regex pattern: https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503/discussions/84#69121093e8b480e709447d5e. This will lead to incorrect tokenization. You should set the `fix_mistral_regex=True` flag when loading this tokenizer to fix this issue.
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:11:49.509530-0700 | reset | INFO - Compression lifecycle reset
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:11:49.510333-0700 | from_modifiers | INFO - Creating recipe from modifiers
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:11:49.555385-0700 | initialize | INFO - Compression lifecycle initialized for 1 modifiers
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:11:49.555665-0700 | IndependentPipeline | INFO - Inferred `DataFreePipeline` for `QuantizationModifier`
chaiml-1007-tl-ads-loving-v11-uploader: Some parameters are on the meta device because they were offloaded to the cpu.
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:12:04.451655-0700 | finalize | INFO - Compression lifecycle finalized for 1 modifiers
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:12:12.595647-0700 | post_process | WARNING - Optimized model is not saved. To save, please provide`output_dir` as input arg.Ex. `oneshot(..., output_dir=...)`
chaiml-1007-tl-ads-loving-v11-uploader: Saving to /dev/shm/model_output...
chaiml-1007-tl-ads-loving-v11-uploader: 2026-03-19T16:12:12.623118-0700 | get_model_compressor | INFO - skip_sparsity_compression_stats set to True. Skipping sparsity compression statistic calculations. No sparsity compressor will be applied.
chaiml-1007-tl-ads-loving-v11-uploader: Cleaning quantization config in /dev/shm/model_output
chaiml-1007-tl-ads-loving-v11-uploader: Pushing to ChaiML/1007-tl-ads-loving-FP8
chaiml-1007-tl-ads-loving-v11-uploader: Checking if ChaiML/1007-tl-ads-loving-FP8 already exists in ChaiML
chaiml-1007-tl-ads-loving-v11-uploader: Creating repo ChaiML/1007-tl-ads-loving-FP8 and uploading /dev/shm/model_output to it
chaiml-1007-tl-ads-loving-v11-uploader: ---------- 2026-03-19 16:12:37 (0:00:00) ----------
chaiml-1007-tl-ads-loving-v11-uploader: Files:   hashed 7/11 (238.2K/13.6G) | pre-uploaded: 0/0 (0.0/13.6G) (+11 unsure) | committed: 0/11 (0.0/13.6G) | ignored: 0
chaiml-1007-tl-ads-loving-v11-uploader: Workers: hashing: 4 | get upload mode: 6 | pre-uploading: 0 | committing: 0 | waiting: 116
chaiml-1007-tl-ads-loving-v11-uploader: ---------------------------------------------------
2026-03-19T23:12:45.279408+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
chaiml-1007-tl-ads-loving-v11-uploader: Processed model ChaiML/1007-tl-ads-loving in 114.111s
chaiml-1007-tl-ads-loving-v11-uploader: creating bucket guanaco-vllm-models
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/BaseUtils.py:56: SyntaxWarning: invalid escape sequence '\.'
chaiml-1007-tl-ads-loving-v11-uploader:   RE_S3_DATESTRING = re.compile('\.[0-9]*(?:[Z\\-\\+]*?)')
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/BaseUtils.py:57: SyntaxWarning: invalid escape sequence '\s'
chaiml-1007-tl-ads-loving-v11-uploader:   RE_XML_NAMESPACE = re.compile(b'^(<?[^>]+?>\s*|\s*)(<\w+) xmlns=[\'"](https?://[^\'"]+)[\'"]', re.MULTILINE)
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/Utils.py:240: SyntaxWarning: invalid escape sequence '\.'
chaiml-1007-tl-ads-loving-v11-uploader:   invalid = re.search("([^a-z0-9\.-])", bucket, re.UNICODE)
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/Utils.py:244: SyntaxWarning: invalid escape sequence '\.'
chaiml-1007-tl-ads-loving-v11-uploader:   invalid = re.search("([^A-Za-z0-9\._-])", bucket, re.UNICODE)
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/Utils.py:255: SyntaxWarning: invalid escape sequence '\.'
chaiml-1007-tl-ads-loving-v11-uploader:   if re.search("-\.", bucket, re.UNICODE):
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/Utils.py:257: SyntaxWarning: invalid escape sequence '\.'
chaiml-1007-tl-ads-loving-v11-uploader:   if re.search("\.\.", bucket, re.UNICODE):
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/S3Uri.py:155: SyntaxWarning: invalid escape sequence '\w'
chaiml-1007-tl-ads-loving-v11-uploader:   _re = re.compile("^(\w+://)?(.*)", re.UNICODE)
chaiml-1007-tl-ads-loving-v11-uploader: /usr/lib/python3/dist-packages/S3/FileLists.py:480: SyntaxWarning: invalid escape sequence '\*'
chaiml-1007-tl-ads-loving-v11-uploader:   wildcard_split_result = re.split("\*|\?", uri_str, maxsplit=1)
chaiml-1007-tl-ads-loving-v11-uploader: Bucket 's3://guanaco-vllm-models/' created
chaiml-1007-tl-ads-loving-v11-uploader: uploading /dev/shm/model_output to s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/recipe.yaml s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/recipe.yaml
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/chat_template.jinja s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/chat_template.jinja
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/model.safetensors.index.json s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/model.safetensors.index.json
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/tokenizer_config.json s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/tokenizer_config.json
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/config.json s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/config.json
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/generation_config.json s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/generation_config.json
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/special_tokens_map.json s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/special_tokens_map.json
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/tokenizer.json s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/tokenizer.json
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/model-00003-of-00003.safetensors s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/model-00003-of-00003.safetensors
2026-03-19T23:13:45.381076+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/model-00001-of-00003.safetensors s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/model-00001-of-00003.safetensors
chaiml-1007-tl-ads-loving-v11-uploader: cp /dev/shm/model_output/model-00002-of-00003.safetensors s3://guanaco-vllm-models/chaiml-1007-tl-ads-loving-v11/default/model-00002-of-00003.safetensors
Job chaiml-1007-tl-ads-loving-v11-uploader completed after 184.69s with status: succeeded
Stopping job with name chaiml-1007-tl-ads-loving-v11-uploader
Pipeline stage VLLMUploader completed in 185.17s
run pipeline stage %s
Running pipeline stage VLLMTemplater
Pipeline stage VLLMTemplater completed in 0.63s
run pipeline stage %s
Running pipeline stage VLLMDeployer
Creating inference service chaiml-1007-tl-ads-loving-v11
Waiting for inference service chaiml-1007-tl-ads-loving-v11 to be ready
2026-03-19T23:14:45.488283+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
2026-03-19T23:15:45.581753+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
2026-03-19T23:16:45.677085+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
Inference service chaiml-1007-tl-ads-loving-v11 ready after 210.60983562469482s
Pipeline stage VLLMDeployer completed in 211.28s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.387892246246338s
Received healthy response to inference request in 2.3398613929748535s
Received healthy response to inference request in 2.3071863651275635s
Received healthy response to inference request in 2.2393300533294678s
Received healthy response to inference request in 2.2401273250579834s
Received healthy response to inference request in 2.431746244430542s
Received healthy response to inference request in 2.415842294692993s
Received healthy response to inference request in 2.3562302589416504s
Received healthy response to inference request in 2.404745101928711s
2026-03-19T23:17:45.777338+00:00 monitor updated for chaiml-1007-tl-ads-loving_v11
Received healthy response to inference request in 2.248886823654175s
Received healthy response to inference request in 2.2992746829986572s
Received healthy response to inference request in 2.2547802925109863s
Received healthy response to inference request in 2.383187770843506s
Received healthy response to inference request in 2.263122320175171s
Received healthy response to inference request in 2.272528648376465s
Received healthy response to inference request in 2.425915479660034s
Received healthy response to inference request in 2.226090431213379s
Received healthy response to inference request in 2.3451473712921143s
Received healthy response to inference request in 2.330859899520874s
Received healthy response to inference request in 2.3160033226013184s
Received healthy response to inference request in 2.334359884262085s
Received healthy response to inference request in 2.4542484283447266s
Received healthy response to inference request in 2.2667412757873535s
Received healthy response to inference request in 2.5366532802581787s
Received healthy response to inference request in 2.3291728496551514s
Received healthy response to inference request in 2.3063528537750244s
Received healthy response to inference request in 2.2674522399902344s
Received healthy response to inference request in 2.2670624256134033s
Received healthy response to inference request in 2.392122507095337s
Received healthy response to inference request in 2.239518880844116s
30 requests
0 failed requests
5th percentile: 2.2394150257110597
10th percentile: 2.240066480636597
20th percentile: 2.261453914642334
30th percentile: 2.267335295677185
40th percentile: 2.3035215854644777
50th percentile: 2.322588086128235
60th percentile: 2.3365604877471924
70th percentile: 2.364317512512207
80th percentile: 2.3946470260620116
90th percentile: 2.426498556137085
95th percentile: 2.4441224455833432
99th percentile: 2.5127558732032775
mean time: 2.3294147650400796
Pipeline stage StressChecker completed in 72.60s
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
Shutdown handler de-registered
chaiml-1007-tl-ads-loving_v11 status is now deployed due to DeploymentManager action
chaiml-1007-tl-ads-loving_v11 status is now inactive due to system request
chaiml-1007-tl-ads-loving_v11 status is now torndown due to DeploymentManager action