Linkornith-aiResource
Ornith-1.5-397B demonstrates that open-weights architectures trained with self-improving agentic loops can compete directly with leading proprietary systems in coding benchmarks, offering teams an unconstrained base model for complex agentic workflows.
Building on Ornith-1.0 (which incorporated Qwen3.5 and Gemma4 checkpoints), Ornith-1.5 replaces fixed human-curated datasets with ongoing self-generated task discovery and RL policy refinement. While deploying the unquantised bf16 model demands enterprise-grade multi-GPU setups (such as 8x H200s for roughly 800 GB memory), standard compatibility with vLLM, SGLang, and OpenAI chat completions ensures straightforward integration into existing coding and agent harnesses.
Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.
Similar in content, found by the AI search.
Link:deep-reinforce.com
Ornith-1.0 is an open-source model family tailored for agentic coding, spanning parameter sizes from 9B to 397B. Built on Gemma 4 and Qwen 3.5, the architecture trains models to jointly generate both code solutions and the task-specific scaffolds directing them.
Article:Luis Chavez-Mattos
Deep Reinforce has introduced Ornith 1.5 35B-A3B, a mixture-of-experts model activating roughly 3 billion parameters per token out of 35 billion. According to published benchmarks, it surpasses direct competitors like Qwen3.6-35B across coding and agentic tasks, while closing the gap on much larger models such as Qwen3.5-397B.
Repository:ornith-ai/Ornith-1
The Ornith-1 repository provides an open-source model family designed for agentic tasks using autonomous self-improvement. Spanning Ornith-1.0 and Ornith-1.5, the project implements self-scaffolding and automated generation of training rollouts for reinforcement learning.
2kstars
Link:Kwaipilot
Kwaipilot has released KAT-Coder-V2.5-Dev, an open-weight Mixture of Experts model based on Qwen3.6-35B-A3B. Featuring 35 billion total and 3 billion activated parameters, it has been specifically post-trained for agentic software engineering tasks.
Link:Poolside
Poolside has released Laguna-S-2.1, a Mixture-of-Experts language model featuring 118B total parameters with approximately 8B activated parameters per token. Targeted at software engineering and long-horizon agentic workflows, the model is available under the permissive OpenMDW-1.1 license.
Link:Z.ai
Z.ai has released GLM-5.2 on Hugging Face under the permissive MIT licence. The flagship open-weights model is designed for long-horizon agentic and coding workflows, featuring a robust 1M-token context window. Architectural updates including IndexShare and an enhanced MTP layer significantly cut computation FLOPs and accelerate speculative decoding.
Parts of this content were created with the assistance of AI.
