smolagents LLM leaderboard
🏆
140
A leaderboard for LLMs powering smolagents
Jędrzej Grabala
jgitsolutions
AI & ML interests
Local Drive Human Overseered System of Agents, LLMs, Langchains & other useful stuff on mid-to-low-end of commercial hardware.
Recent Activity
liked
a Space
about 4 hours ago
smolagents/smolagents-leaderboard
liked
a Space
about 12 hours ago
microsoft/TRELLIS.2
liked
a model
1 day ago
distilbert/distilbert-base-uncased
Organizations
liked
a
Space
about 4 hours ago
liked
a
Space
about 12 hours ago
liked
a
model
1 day ago
liked
a
Space
19 days ago
liked
a
Space
21 days ago
reacted to
branikita's
post with 🚀
24 days ago
Post
1870
We've published a comprehensive evaluation of the Feetech STS3250 servo actuator.
Key Findings:
- Speed: 77.6 RPM (exceeds spec by 3.2%)
- Backlash: 0.43° (within 0.5° limit)
- Repeatability: ±0.02mm at 95mm radius
- Peak torque: 48 kg·cm
- Sustained torque: ~25 kg·cm after thermal protection
Full review: https://robonine.com/feetech-sts3250-smart-actuator-evaluation-of-accuracy-torque-and-backlash/
Key Findings:
- Speed: 77.6 RPM (exceeds spec by 3.2%)
- Backlash: 0.43° (within 0.5° limit)
- Repeatability: ±0.02mm at 95mm radius
- Peak torque: 48 kg·cm
- Sustained torque: ~25 kg·cm after thermal protection
Full review: https://robonine.com/feetech-sts3250-smart-actuator-evaluation-of-accuracy-torque-and-backlash/
- 1 reply
updated
a
Space
29 days ago
liked
2
Spaces
about 1 month ago
liked
a
Space
about 2 months ago
reacted to
salma-remyx's
post with 👍
3 months ago
Post
3243
Reproducing research code shouldn't take longer than reading the paper.
For papers that include code, setting up the right environment often means hours of dependency hell and configuration debugging.
At Remyx AI, we built an agent that automatically creates and tests Docker images for research papers, then shares them publicly so anyone can reproduce results with a single command.
We just submitted PR #908 to integrate this directly into arXiv Labs.
If you believe in making reproducible research accessible to everyone, give it a bump!: https://github.com/arXiv/arxiv-browse/pull/908
For papers that include code, setting up the right environment often means hours of dependency hell and configuration debugging.
At Remyx AI, we built an agent that automatically creates and tests Docker images for research papers, then shares them publicly so anyone can reproduce results with a single command.
We just submitted PR #908 to integrate this directly into arXiv Labs.
If you believe in making reproducible research accessible to everyone, give it a bump!: https://github.com/arXiv/arxiv-browse/pull/908
- 3 replies
liked
3
Spaces
6 months ago
liked
4
Spaces
8 months ago
reacted to
fdaudens's
post with 🔥
8 months ago
Post
3214
Forget everything you know about transcription models - NVIDIA's parakeet-tdt-0.6b-v2 changed the game for me!
Just tested it with Steve Jobs' Stanford speech and was speechless (pun intended). The video isn’t sped up.
3 things that floored me:
- Transcription took just 10 seconds for a 15-min file
- Got a CSV with perfect timestamps, punctuation & capitalization
- Stunning accuracy (correctly captured "Reed College" and other specifics)
NVIDIA also released a demo where you can click any transcribed segment to play it instantly.
The improvement is significant: number 1 on the ASR Leaderboard, 6% error rate (best in class) with complete commercial freedom (cc-by-4.0 license).
Time to update those Whisper pipelines! H/t @Steveeeeeeen for the finding!
Model: nvidia/parakeet-tdt-0.6b-v2
Demo: nvidia/parakeet-tdt-0.6b-v2
ASR Leaderboard: hf-audio/open_asr_leaderboard
Just tested it with Steve Jobs' Stanford speech and was speechless (pun intended). The video isn’t sped up.
3 things that floored me:
- Transcription took just 10 seconds for a 15-min file
- Got a CSV with perfect timestamps, punctuation & capitalization
- Stunning accuracy (correctly captured "Reed College" and other specifics)
NVIDIA also released a demo where you can click any transcribed segment to play it instantly.
The improvement is significant: number 1 on the ASR Leaderboard, 6% error rate (best in class) with complete commercial freedom (cc-by-4.0 license).
Time to update those Whisper pipelines! H/t @Steveeeeeeen for the finding!
Model: nvidia/parakeet-tdt-0.6b-v2
Demo: nvidia/parakeet-tdt-0.6b-v2
ASR Leaderboard: hf-audio/open_asr_leaderboard
- 1 reply
liked
a
Space
8 months ago