developer_uid: chai_backend_admin
submission_id: chaiml-gspo-glm47-chai-_75693_v1
model_name: chaiml-gspo-glm47-chai-_75693_v1
model_group: ChaiML/gspo-glm47-chai-r
status: torndown
timestamp: 2026-03-20T06:11:43+00:00
num_battles: 9037
num_wins: 5212
celo_rating: 1348.85
family_friendly_score: 0.0
family_friendly_standard_error: 0.0
submission_type: basic
model_repo: ChaiML/gspo-glm47-chai-rm-zy-data-step500
model_architecture: Glm4MoeForCausalLM
model_num_parameters: 24110003200.0
best_of: 8
max_input_tokens: 1500
max_output_tokens: 80
reward_model: default
display_name: chaiml-gspo-glm47-chai-_75693_v1
ineligible_reason: max_output_tokens!=64
is_internal_developer: True
language_model: ChaiML/gspo-glm47-chai-rm-zy-data-step500
model_size: 24B
ranking_group: single
us_pacific_date: 2026-03-16
win_ratio: 0.5767400686068386
generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.0, 'top_k': 60, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['####', '</s>', '<|im_end|>', '<|assistant|>', '<|user|>'], 'max_input_tokens': 1500, 'best_of': 8, 'max_output_tokens': 80}
formatter: {'memory_template': "[gMASK]<sop><|system|>\n{bot_name}'s persona: {memory}", 'prompt_template': '', 'bot_template': '<|assistant|>\n{bot_name}: {message}', 'user_template': '<|user|>\n{message}', 'response_template': '<|assistant|>\n<think></think>\n{bot_name}:', '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 VLLMUploader
Starting job with name chaiml-gspo-glm47-chai-75693-v1-uploader
Waiting for job on chaiml-gspo-glm47-chai-75693-v1-uploader to finish
chaiml-gspo-glm47-chai-75693-v1-uploader: Using quantization_mode: fp8
chaiml-gspo-glm47-chai-75693-v1-uploader: Checking if ChaiML/gspo-glm47-chai-rm-zy-data-step500-FP8 already exists in ChaiML
chaiml-gspo-glm47-chai-75693-v1-uploader: Downloading snapshot of ChaiML/gspo-glm47-chai-rm-zy-data-step500...
2026-03-16T23:50:34.690359+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
2026-03-16T23:51:43.913902+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
2026-03-16T23:52:43.998796+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
2026-03-16T23:53:44.088023+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
chaiml-gspo-glm47-chai-75693-v1-uploader: Downloaded in 240.134s
chaiml-gspo-glm47-chai-75693-v1-uploader: Loading /tmp/model_input...
chaiml-gspo-glm47-chai-75693-v1-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-gspo-glm47-chai-75693-v1-uploader: `torch_dtype` is deprecated! Use `dtype` instead!
2026-03-16T23:54:44.184112+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
chaiml-gspo-glm47-chai-75693-v1-uploader: Some parameters are on the meta device because they were offloaded to the cpu.
chaiml-gspo-glm47-chai-75693-v1-uploader: Applying quantization...
chaiml-gspo-glm47-chai-75693-v1-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-gspo-glm47-chai-75693-v1-uploader: 2026-03-16T16:55:15.605006-0700 | reset | INFO - Compression lifecycle reset
chaiml-gspo-glm47-chai-75693-v1-uploader: 2026-03-16T16:55:15.672242-0700 | from_modifiers | INFO - Creating recipe from modifiers
2026-03-16T23:55:44.268121+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
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-gspo-glm47-combi-53542-v1-uploader
Waiting for job on chaiml-gspo-glm47-combi-53542-v1-uploader to finish
chaiml-gspo-glm47-chai-75693-v1-uploader: 2026-03-16T16:56:11.722927-0700 | initialize | INFO - Compression lifecycle initialized for 1 modifiers
chaiml-gspo-glm47-chai-75693-v1-uploader: 2026-03-16T16:56:11.723402-0700 | IndependentPipeline | INFO - Inferred `DataFreePipeline` for `QuantizationModifier`
chaiml-gspo-glm47-combi-53542-v1-uploader: Using quantization_mode: fp8
2026-03-16T23:56:44.362627+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
chaiml-gspo-glm47-chai-75693-v1-uploader: Some parameters are on the meta device because they were offloaded to the cpu.
chaiml-gspo-glm47-combi-53542-v1-uploader: Checking if ChaiML/gspo-glm47-combine-rm82-zy-data-step450-FP8 already exists in ChaiML
chaiml-gspo-glm47-combi-53542-v1-uploader: Downloading snapshot of ChaiML/gspo-glm47-combine-rm82-zy-data-step450...
2026-03-16T23:57:02.674036+00:00 monitor updated for chaiml-gspo-glm47-combi_53542_v1
2026-03-16T23:57:44.496406+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
2026-03-16T23:58:02.804320+00:00 monitor updated for chaiml-gspo-glm47-combi_53542_v1
2026-03-16T23:58:44.633721+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
2026-03-16T23:59:02.947940+00:00 monitor updated for chaiml-gspo-glm47-combi_53542_v1
2026-03-16T23:59:44.772893+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
2026-03-17T00:00:03.090793+00:00 monitor updated for chaiml-gspo-glm47-combi_53542_v1
chaiml-gspo-glm47-combi-53542-v1-uploader: Downloaded in 239.011s
2026-03-17T00:00:44.919968+00:00 monitor updated for chaiml-gspo-glm47-chai-_75693_v1
chaiml-gspo-glm47-combi-53542-v1-uploader: Loading /tmp/model_input...
chaiml-gspo-glm47-combi-53542-v1-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-gspo-glm47-combi-53542-v1-uploader: `torch_dtype` is deprecated! Use `dtype` instead!
chaiml-gspo-glm47-chai-_75693_v1 status is now deployed due to DeploymentManager action
chaiml-gspo-glm47-chai-_75693_v1 status is now inactive due to system request
chaiml-gspo-glm47-chai-_75693_v1 status is now torndown due to DeploymentManager action