pytorch

non-profit

https://pytorch.org/

AI & ML interests

An open source machine learning framework that accelerates the path from research prototyping to production deployment.

posted an update 13 days ago

Post

314

Uhh did Opus 4.8 cheat on PostTrainBench??

it found an API key in the PostTrainBench environment that allowed it to generate synthetic training data without using GPU hours, boosting the base model by 0.4913

Source: https://posttrainbench.com/traces/run.html?id=claude_non_api_max_claude-opus-4-8_10h_run1__healthbench_Qwen_Qwen3-4B-Base_17315102#tab=trace

1 reply

·

updated 2 models 4 months ago

pytorch/gemma-3-12b-it-INT4

3B • Updated Mar 11 • 6.48k • 1

pytorch/gemma-3-27b-it-INT4

6B • Updated Mar 11 • 6.3k

in pytorch/gemma-3-27b-it-AWQ-INT4 5 months ago

How are you using the checkpoint?

#2 opened 5 months ago by

in pytorch/gemma-3-27b-it-AWQ-INT4 6 months ago

Adding `safetensors` variant of this model

#1 opened 9 months ago by

updated a collection 7 months ago

TorchAO PARQ models

Models trained with PARQ: https://arxiv.org/abs/2503.15748 using TorchAO • 3 items • Updated Dec 17, 2025

updated 3 models 7 months ago

pytorch/Phi-4-mini-instruct-parq-3w-4e-shared

Text Generation • Updated Dec 16, 2025 • 101

pytorch/Phi-4-mini-instruct-parq-2w-4e-shared

Text Generation • Updated Dec 16, 2025 • 126 • 1

pytorch/Phi-4-mini-instruct-parq-4w-4e-shared-gsm

Text Generation • Updated Dec 16, 2025 • 101

lvj

updated 3 models 7 months ago

pytorch/Phi-4-mini-instruct-parq-3w-4e-shared

Text Generation • Updated Dec 16, 2025 • 101

pytorch/Phi-4-mini-instruct-parq-2w-4e-shared

Text Generation • Updated Dec 16, 2025 • 126 • 1

pytorch/Phi-4-mini-instruct-parq-4w-4e-shared-gsm

Text Generation • Updated Dec 16, 2025 • 101

lvj

updated a collection 7 months ago

TorchAO Quantized Phi-4-mini-instruct

TorchAO quantized Phi-4-mini-instruct models from PyTorch team, runnable in A100, H100 through vLLM and in mobile devices through ExecuTorch • 7 items • Updated Dec 16, 2025 • 3

updated a collection 7 months ago

TorchAO QAT models

models quantized through quantization aware training from torchao • 2 items • Updated Dec 8, 2025

authored a paper 8 months ago

Gradio: Hassle-Free Sharing and Testing of ML Models in the Wild

Paper • 1906.02569 • Published Jun 6, 2019 • 2