ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
Paper
•
2504.11536
•
Published
•
33
Victor Mustar PRO
victor
AI & ML interests
Building the UX of this website
Recent Activity
upvoted
a
paper
about 14 hours ago
ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
liked
a model
about 18 hours ago
lmstudio-community/granite-3.3-2b-instruct-GGUF
updated
a Space
about 18 hours ago
victor/spaces-trending
Organizations
victor's activity
upvoted
a
paper
about 14 hours ago
liked
a
model
about 18 hours ago
updated
a
Space
about 18 hours ago
liked
a
model
about 21 hours ago
liked
a
model
about 23 hours ago
How to run 🐳 DeepSite locally
16
#74 opened 1 day ago
by
enzostvs
upvoted
2
papers
1 day ago
liked
a
Space
1 day ago
reacted to
nyuuzyou's
post with 👍
1 day ago
Post
5407
🇷🇺 Russian Forum Messages Dataset -
nyuuzyou/ruforum
Collection of approximately 58 million Russian forum messages featuring:
- Complete message content from Russian online forums spanning 2010-2025
- Comprehensive metadata including unique message IDs and timestamps
- Full text content preserving original user discussions and interactions
- Monolingual dataset focused exclusively on Russian language content
This dataset offers a unique textual archive of Russian online conversations suitable for text generation, sentiment analysis, and language modeling research. Released to the public domain under CC0 1.0 license.
Collection of approximately 58 million Russian forum messages featuring:
- Complete message content from Russian online forums spanning 2010-2025
- Comprehensive metadata including unique message IDs and timestamps
- Full text content preserving original user discussions and interactions
- Monolingual dataset focused exclusively on Russian language content
This dataset offers a unique textual archive of Russian online conversations suitable for text generation, sentiment analysis, and language modeling research. Released to the public domain under CC0 1.0 license.
reacted to
AdinaY's
post with ❤️
1 day ago
Post
3098
🔥 New reasoning models from the Chinese community, by Skywork 天工-昆仑万维
Skywork/skywork-or1-67fa1bcb41b436ef2def76b9
✨Skywork OR1-Math-7B > Optimized for math reasoning
✨Skywork-OR1-7B-preview > Excels in math & coding
✨Skywork-OR1-32B-preview > Matches Deepseek-R1 on math (AIME24/25) and coding (LiveCodeBench)
Released under the Apache 2.0 license 🥳
Final version coming in 2 weeks!
Skywork/skywork-or1-67fa1bcb41b436ef2def76b9
✨Skywork OR1-Math-7B > Optimized for math reasoning
✨Skywork-OR1-7B-preview > Excels in math & coding
✨Skywork-OR1-32B-preview > Matches Deepseek-R1 on math (AIME24/25) and coding (LiveCodeBench)
Released under the Apache 2.0 license 🥳
Final version coming in 2 weeks!
reacted to
thomwolf's
post with 🚀
1 day ago
Post
4130
If you've followed the progress of robotics in the past 18 months, you've likely noticed how robotics is increasingly becoming the next frontier that AI will unlock.
At Hugging Face—in robotics and across all AI fields—we believe in a future where AI and robots are open-source, transparent, and affordable; community-built and safe; hackable and fun. We've had so much mutual understanding and passion working with the Pollen Robotics team over the past year that we decided to join forces!
You can already find our open-source humanoid robot platform Reachy 2 on the Pollen website and the Pollen community and people here on the hub at
pollen-robotics
We're so excited to build and share more open-source robots with the world in the coming months!
At Hugging Face—in robotics and across all AI fields—we believe in a future where AI and robots are open-source, transparent, and affordable; community-built and safe; hackable and fun. We've had so much mutual understanding and passion working with the Pollen Robotics team over the past year that we decided to join forces!
You can already find our open-source humanoid robot platform Reachy 2 on the Pollen website and the Pollen community and people here on the hub at
pollen-robotics
We're so excited to build and share more open-source robots with the world in the coming months!
- 1 reply
reacted to
bartowski's
post with 👍
1 day ago
Post
6272
Access requests enabled for latest GLM models
While a fix is being implemented (https://github.com/ggml-org/llama.cpp/pull/12957) I want to leave the models up for visibility and continued discussion, but want to prevent accidental downloads of known broken models (even though there are settings that could fix it at runtime for now)
With this goal, I've enabled access requests. I don't really want your data, so I'm sorry that I don't think there's a way around that? But that's what I'm gonna do for now, and I'll remove the gate when a fix is up and verified and I have a chance to re-convert and quantize!
Hope you don't mind in the mean time :D
While a fix is being implemented (https://github.com/ggml-org/llama.cpp/pull/12957) I want to leave the models up for visibility and continued discussion, but want to prevent accidental downloads of known broken models (even though there are settings that could fix it at runtime for now)
With this goal, I've enabled access requests. I don't really want your data, so I'm sorry that I don't think there's a way around that? But that's what I'm gonna do for now, and I'll remove the gate when a fix is up and verified and I have a chance to re-convert and quantize!
Hope you don't mind in the mean time :D
reacted to
prithivMLmods's
post with 👍
1 day ago
Post
2153
Try out the demo for Multimodal OCR featuring the implementation of models including
🤗Multimodal OCR Space : prithivMLmods/Multimodal-OCR
📦The models implemented in this Space are:
+ Qwen2VL OCR : prithivMLmods/Qwen2-VL-OCR-2B-Instruct
+ Qwen2VL OCR2 : prithivMLmods/Qwen2-VL-OCR2-2B-Instruct
+ RolmOCR : reducto/RolmOCR
RolmOCR and Qwen2VL OCR. The use case showcases image-text-to-text conversion and video understanding support for the RolmOCR model ! 🚀🤗Multimodal OCR Space : prithivMLmods/Multimodal-OCR
📦The models implemented in this Space are:
+ Qwen2VL OCR : prithivMLmods/Qwen2-VL-OCR-2B-Instruct
[ or ]+ Qwen2VL OCR2 : prithivMLmods/Qwen2-VL-OCR2-2B-Instruct
+ RolmOCR : reducto/RolmOCR
Qwen2VL OCR supports only image-text-to-text in the space.
reacted to
luigi12345's
post with 🚀
1 day ago
Post
1805
BREAKING NEWS! 🚀 OpenAI’s GPT-4.1 API Models Are Here – Built for Developers
OpenAI has launched GPT-4.1, GPT-4.1 Mini, and GPT-4.1 Nano—models engineered for real-world coding, instruction following, and long-context tasks. 
🔧 Key Dev Features
• Coding Performance: GPT-4.1 scores 54.6% on SWE-bench Verified, outperforming GPT-4o by 21.4% and GPT-4.5 by 26.6%. It handles diffs more precisely, reduces unnecessary edits, and adheres to formatting constraints. 
• Long Context: All models support up to 1 million tokens—8x more than GPT-4o—enabling full repo analysis and deep document comprehension. 
• Instruction Following: Improved multi-step reasoning and formatting accuracy, with a 10.5% gain over GPT-4o on MultiChallenge. 
• Latency & Cost: GPT-4.1 is 40% faster and 80% cheaper per query than GPT-4o. Mini and Nano versions offer even greater speed and affordability. 
🧠 Model Lineup
Model Context Window Use Case Cost per 1M Tokens
GPT-4.1 1M tokens Production-grade coding & agents $2.00 input / $8.00 output
GPT-4.1 Mini 1M tokens Balanced performance, cost-sensitive apps $0.40 / $1.60
GPT-4.1 Nano 1M tokens Ultra-fast, lightweight tasks $0.10 / $0.40
🛠️ Access & Tools
• API Only: Available via OpenAI API and Playground—ChatGPT remains on GPT-4o. 
• Prompting Guide: Optimized prompts for agentic coding workflows. 
• Benchmarks & Pricing: Detailed comparisons and cost breakdowns. 
For more information, [visit the official announcement](https://openai.com/index/gpt-4-1)
OpenAI has launched GPT-4.1, GPT-4.1 Mini, and GPT-4.1 Nano—models engineered for real-world coding, instruction following, and long-context tasks. 
🔧 Key Dev Features
• Coding Performance: GPT-4.1 scores 54.6% on SWE-bench Verified, outperforming GPT-4o by 21.4% and GPT-4.5 by 26.6%. It handles diffs more precisely, reduces unnecessary edits, and adheres to formatting constraints. 
• Long Context: All models support up to 1 million tokens—8x more than GPT-4o—enabling full repo analysis and deep document comprehension. 
• Instruction Following: Improved multi-step reasoning and formatting accuracy, with a 10.5% gain over GPT-4o on MultiChallenge. 
• Latency & Cost: GPT-4.1 is 40% faster and 80% cheaper per query than GPT-4o. Mini and Nano versions offer even greater speed and affordability. 
🧠 Model Lineup
Model Context Window Use Case Cost per 1M Tokens
GPT-4.1 1M tokens Production-grade coding & agents $2.00 input / $8.00 output
GPT-4.1 Mini 1M tokens Balanced performance, cost-sensitive apps $0.40 / $1.60
GPT-4.1 Nano 1M tokens Ultra-fast, lightweight tasks $0.10 / $0.40
🛠️ Access & Tools
• API Only: Available via OpenAI API and Playground—ChatGPT remains on GPT-4o. 
• Prompting Guide: Optimized prompts for agentic coding workflows. 
• Benchmarks & Pricing: Detailed comparisons and cost breakdowns. 
For more information, [visit the official announcement](https://openai.com/index/gpt-4-1)
reacted to
neph1's
post with 🚀
1 day ago
Post
2113
I know Hunyuan Video is yesterday's jam, but in case you're looking for some cinematic LoRA's (and don't like civitai for some reason), I've uploaded my most popular ones to hf. They are:
1980s fantasy: neph1/1980s_Fantasy_Movies_Hunyuan_Video_Lora
1950s scifi: neph1/50s_scifi_hunyuan_video_lora
1920s horror: neph1/1920s_horror_hunyuan_video_lora
1980s fantasy: neph1/1980s_Fantasy_Movies_Hunyuan_Video_Lora
1950s scifi: neph1/50s_scifi_hunyuan_video_lora
1920s horror: neph1/1920s_horror_hunyuan_video_lora
reacted to
davidberenstein1957's
post with 👀
1 day ago
Post
1141
RealHarm: A Collection of Real-World Language Model Application Failure
I'm David from Giskard, and we work on securing your Agents.
Today, we are launching RealHarm: a dataset of real-world problematic interactions with AI agents, drawn from publicly reported incidents.
Check out the dataset and paper: https://realharm.giskard.ai/
I'm David from Giskard, and we work on securing your Agents.
Today, we are launching RealHarm: a dataset of real-world problematic interactions with AI agents, drawn from publicly reported incidents.
Check out the dataset and paper: https://realharm.giskard.ai/
reacted to
Yehor's
post with 🚀
1 day ago
Post
2032
Made a workable program that uses IREE runtime using Rust to inference wav2vec2-bert model for Automatic Speech Recognition.
updated
a
Space
3 days ago
published
a
Space
3 days ago