The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestDeepreinforce Ai1.0
Deepreinforce AiText / LLM

DeepReinforce's Ornith-1.0-9B Targets Agentic Coding

A compact, MIT-licensed 9B model built for autonomous coding tasks arrives on Hugging Face.

Jun 21, 2026
UpdateMIT
Ornith-1.5-122B

DeepReinforce has released Ornith-1.0-9B, a nine-billion-parameter model aimed squarely at agentic coding workflows. It is a dense (non–mixture-of-experts) model distributed under the permissive MIT license, which allows both commercial and research use with minimal restrictions.

The pitch is size. At 9B parameters, Ornith sits in a sweet spot for teams that want a model small enough to run on modest hardware but capable enough to handle code generation and multi-step, tool-using tasks. That framing puts it in a growing category of compact coding models designed to power autonomous agents rather than one-shot completions.

Why it matters

Agentic coding — where a model plans, writes, executes, and revises code across several steps — has become a defining use case for open-weight models. A few reasons this release is worth noting:

  • Permissive licensing: MIT terms make it straightforward to embed in products or fine-tune further.
  • Deployability: A 9B footprint is friendlier to local and cost-sensitive deployments than frontier-scale systems.
  • Focus: The model is positioned specifically for code and agent tasks rather than as a general-purpose chatbot.

DeepReinforce has not published detailed benchmark figures or a context-length specification in this record, so real-world evaluation will depend on independent testing. For now, the weights and details are live on Hugging Face, and the company outlines the release on its own project page. As the first entry in the Ornith line, version 1.0 sets a baseline the team will presumably build on.

Sources

  • deepreinforce-ai/Ornith-1.0-9B

    Hugging Face

    Visit
OlderOrnith-1.0-35B brings a mid-size MoE to agentic codingDeepreinforce Ai · Text / LLM · 5 months agoNewerAgents-A1: A 35B MoE Built for Agentic ScalingInternScience · Reasoning · 5 months ago

Get the model

Hugging Face

Specs

Parameters9B
Context window262K tokens
Size18.8 GB
PrecisionBF16
ArchitectureQwen3_5ForConditionalGeneration
LicenseMIT
Downloads186.6K
Likes262

Can you run it?

Runs on just about anything.

  • BF16 (as published)

    24 GB GPU (RTX 3090 / 4090) · 32 GB Mac

    20.9 GB
  • 8-bit

    Any modern laptop · 8 GB RAM

    2.1 GB
  • 4-bit

    Any modern laptop · 8 GB RAM

    2.1 GB
Your machine
GGUF builds
  • KV cache sized for 8,192 tokens of context; longer prompts need proportionally more.

Based on Hugging Face weights metadata. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

CodeText / LLM
7 versions — view changelog

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More from Deepreinforce Ai

All Deepreinforce Ai releases →
Ornith-1.5-122B
Deepreinforce Ai/Text / LLMConsumer GPU

DeepReinforce Releases Ornith 1.0, a 35B Reasoning Model

The new dense model ships in GGUF format under a permissive MIT license, aimed at local and self-hosted deployment.

Jun 25, 2026
Ornith-1.5-122B
Deepreinforce Ai/Text / LLMMulti-GPU server

DeepReinforce debuts Ornith-1.0, a 397B MoE model

The flagship of a new open model family arrives under a permissive MIT license, with reasoning among its stated strengths.

Jun 23, 2026
Ornith-1.5-122B
Deepreinforce Ai/Text / LLMRuns anywhere

Ornith-1.0-35B brings a mid-size MoE to agentic coding

An MIT-licensed mixture-of-experts model targets self-scaffolding code tasks without the footprint of a frontier system.

Jun 21, 2026

More in Text / LLM

All Text / LLM →
Meta AI/Text / LLM

Meta AI debuts Muse Ember, a compact open model

The company's smallest, most efficient open-weight language model targets lean, on-device text workloads.

Nov 10, 2026
Inkling-2-Small
Thinkingmachines/Vision-LanguageMulti-GPU server

Thinking Machines ships Inkling-2-Small, an open MoE VLM

The Apache-2.0 vision-language model uses a mixture-of-experts design and lands openly on Hugging Face.

Nov 4, 2026
LongCat 2.5
Meituan/Text / LLMDatacenter

Meituan Releases LongCat 2.5, an MIT-Licensed MoE

The Chinese delivery giant expands its open-model push with a multimodal, agentic mixture-of-experts system.

Nov 4, 2026