Macaron V1 adds LoRA RL on GLM 5.2 and claims SOTA benchmark gains
Xianbao_QIAN · x · 2026-07-22
Macaron V1, built on top of the GLM 5.2 base model with LoRA RL, is being presented as an early example of how much potential GLM 5.2 has.
- The release claims strong benchmark gains and shows several state-of-the-art results.
- It uses specialist LoRAs trained with the Mind Lab Toolkit.
- The stack is served through a Mix-of-LoRA harness.
- The model is open-weight under the MIT license.
- A blog post and hosted API are available.
More from Models
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11
- awesome-llm-leaderboards: an open-source directory of LLM leaderboards, pricing tables, comparison tools — Last_Establishment_1 · 2026-09-11
- Anthropic claims it works to keep eval environments unidentifiable to models — MaxKannen · 2026-09-11
- Nex N2.5 Pro released on Hugging Face with 407GB of weights — jinnyjuice · 2026-09-11
- RoMa v2 image matching model unveiled in the usual black poster — ducha_aiki · 2026-09-11
- OpenAI rated Astra 'Critical' for cyber capabilities — and admits it's harder to monitor — theguywhobuilds · 2026-09-11