RequestHunt is a community-driven platform that collects and curates feature requests from X (Twitter), Reddit, and GitHub. We help product teams discover what features real users are asking for, making it easier to validate product ideas and prioritize feature development based on actual user demand.

How does RequestHunt collect feature requests?

RequestHunt uses AI-powered extraction to automatically collect feature requests from social media platforms including X, Reddit, and GitHub. The platform monitors discussions, issues, and posts to identify genuine user feature requests, extracting key information like the requested feature, product mentioned, and user sentiment.

Who uses RequestHunt?

RequestHunt is used by product managers, founders, developers, and product teams who want to discover what features real users are requesting. It's particularly valuable for startups looking to validate ideas, established companies prioritizing their roadmap, and developers building products that solve real user problems.

Is RequestHunt free to use?

Yes, browsing and searching feature requests on RequestHunt is completely free. We also offer API access for developers who want to integrate feature request data into their own applications. Check our pricing page for API plan details.

How often is RequestHunt data updated?

RequestHunt continuously collects feature requests from Reddit, X (Twitter), and GitHub. You can also trigger on-demand scraping via the API or web interface to get the freshest data for any topic.

[trl] Multi-GPU sampling for vLLM in GRPO Trainer

Allow specifying multiple GPUs for vLLM device in GRPO Trainer, instead of being limited to a single GPU. This would improve RL performance by removing the single GPU bottleneck.

Original Source

@@nch0w | 19 pts | 1/30/2025

### Feature request It seems that the vLLM device can only be set in [GRPOConfig.vllm_device](https://github.com/huggingface/trl/blob/main/trl/trainer/grpo_config.py#L130), which is a string corresponding to a CUDA device identifier. I think this implies that the vLLM device can only use a single GPU, which can be a bottleneck for RL. It is also possible to use a subset by setting the CUDA_VISIBLE_DEVICE environment variable, but this might break TRL. Is there a more convenient way to specify multiple GPUs in a single node for training (or any hacks that would work now)? Furthermore, there might need to be more detailed configurations for multi-node vLLM/GPRO training runs. ### Motivation Enhance training efficiency for RL with >single GPU sampling. ### Your contribution N/A

Discussion

Loading comments...