merve
merve is I'm gepa-ing again after few fixes, in case you want to try though I share the project based on TRL and OpenEnv for your agents. It is ranked #1176 on the AI Radar, in AI agents, first seen 13 days ago and shared in 4 posts (81.5k views).
What merve says about itself
Storage bucket merve/blender-grpo-rollouts on Hugging Face
merve/blender-grpo-rollouts - Storage Bucket Hugging Face Log In Sign Up Buckets: merve / blender-grpo-rollouts Files xet 947 MB 3,319 files Updated 12 days ago ⌘ K Name Size Uploaded Xet hash low-poly-rooms-gepa-short 13 days ago 106 items low-poly-rooms-gepa-short-20260906-115329 13 days ago 253 items low-poly-rooms-gepa-short-20260906-131000 13 days ago 185 items low-poly-rooms-gepa-short-20260906-164601 13 days ago 3 items low-poly-rooms-gepa-short-20260906-164830 13 days ago 136 items low-poly-rooms-gepa-short-20260906-172230 13 days ago 69 items low-poly-rooms-gepa-short-20260906-203544 12 days ago 50 items…
What people said about merve on X
I spent my Astra tokens trying to teach small Qwen3.5 (9B) to create lowpoly 3D rooms using Blender 🏡 you (or Astra) can take this challenge today https://huggingface.co/buckets/merve/blender-grpo-gepa here's what I learned: > you need to align geometry with deterministic rewards and VLM-as-a-judge (GLM-5.3-Flash…
— @mervenoyann, 12 days ago · 316 likes · see the post
I'm doing best of both worlds by training Qwen3.5-9B to do isometric low poly worlds, with GLM 5.3 Flash as judge currently running gepa with different rewards, initially I couldn't get any renders, now the ones I get aren't semantically good then we do grpo watch my rollouts here…
— @mervenoyann, 13 days ago · 134 likes · see the post
I'm gepa-ing again after few fixes, in case you want to try though I share the project based on TRL and OpenEnv for your agents https://huggingface.co/buckets/merve/blender-grpo-gepa
— @mervenoyann, 13 days ago · 61 likes · see the post
Alternatives to merve
- muse.ai — New connectors are live today. Come build with us.
- academy.dair.ai — Chat with Paper
- alphaXiv — τ-BENCH is introduced as a benchmark evaluating AI systems on their ability to construct deployable, customer-facing…
- Sakana Fugu — One model to command them all
- typesafe.ai — yes - we're hiring!
- context.dev — powers the branding, company context, and search.
merve in numbers
- Rank on the AI Radar: #1176 of 1190
- Shared in 4 posts by 1 account: @mervenoyann
- 81.5k views on those posts
- First seen 13 days ago, last shared 12 days ago
- Pricing seen by Jev: free
- Market: AI agents
FAQ
What is merve?
I'm gepa-ing again after few fixes, in case you want to try though I share the project based on TRL and OpenEnv for your agents It was first shared on X 13 days ago and is ranked #1176 on the AI Radar.
Is merve free?
Yes, it is free to use.
Who shared merve?
1 account on X, including @mervenoyann, in 4 posts totalling 81.5k views.
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 27.2k posts from 4.8k X accounts over the last 14 days, 1.2k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-19 18:08 UTC. Full method.