Open model family (9B dense, 35B MoE, 397B MoE) trained with a self-improvement loop for reasoning, coding and agentic tasks.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Ornith-1.5 is a family of open models in three sizes: 397B MoE, 35B MoE (about 3B active parameters per token) and 9B dense. It extends Ornith-1.0, which was built on Qwen3.5 and Gemma 4 with continued pretraining, mid-training and post-training.
Its training uses a self-improvement loop. The model proposes progressively harder tasks, generates task-specific scaffolds (instructions, tools, decomposition, orchestration) and produces solution rollouts for reinforcement learning. Reward is propagated across all three stages and combines task validity, frontier difficulty (target success rate 0.2) and novelty.
The vendor reports Ornith-1.5-397B at 86.1 on Terminal-Bench 2.1 and 56.0 on DeepSWE, compared with Claude Opus 4.8 (85.0 and 59.0), GLM-5.2 (82.7 and 46.2) and DeepSeek-V4-Flash-0731 (82.7 and 54.4). The 35B scores 68.5 on Terminal-Bench 2.1 and 79.0 on SWE-Bench Verified. The 9B scores 47.0 and 70.6, and a quantized 9B-Mobile version targets iPhone and Android. These are vendor-reported figures and have not been independently verified here. Creators also mention GGUF releases. Licence, download links and vendor identity were not found in the provided material.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Open model family (9B dense, 35B MoE, 397B MoE) trained with a self-improvement loop where the system proposes problems, builds scaffolds and generates RL data. The largest beats GLM 5.2 and nears Opus 4.8 on agentic benchmarks; GGUFs released.
Closest neighbours by meaning, not just by category.
Model
Open-weights family of agentic coding models (9B to 397B) that learns to generate its own task-specific scaffolds.
Model · Accio Lab
Open-weights 35B MoE (3B active) agentic model from Accio Lab, post-trained from Qwen3.6-35B-A3B.
Model · Alibaba
Alibaba's open-weights 2.4T-parameter MoE language model (95B active) from the Qwen3.8 family, with a hosted Max version.
Model · Z.ai
Z.ai's open-weights 753B MoE language model, a post-trained GLM-5.2 aimed at coding, agentic tasks and cybersecurity.
Model · MiniMax
Open-weights native multimodal MoE model from MiniMax, ~428B parameters (~23B active) with 1M-token context.
Model · Meituan
Meituan's open-weights 1.6T-parameter multimodal MoE model (48B active) for agentic coding and visual tasks.