Ornith 1.0 Models by DeepReinforce
Details
Capabilities
Links
WebsiteAbout
Ornith 1.0 is DeepReinforce's first open-source family of agentic coding models, trained with a novel self-scaffolding reinforcement learning approach: rather than using a fixed hand-engineered execution harness, the model jointly proposes both the scaffold and the solution during RL training, then iteratively refines both. The family announcement names four variants: 9B Dense, 31B Dense, 35B MoE, and 397B MoE. Public Hugging Face repos are verified for 9B, 35B, and 397B; the 31B variant was not publicly accessible as of June 27, 2026. Models are MIT-licensed and support OpenAI-compatible tool calls for agent frameworks. Released June 25, 2026.
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs coding, 262k context, and 9B parameters.
Use when the workload needs coding, 262k context, and 35B parameters.
Use when the workload needs coding, 262k context, and 397B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Ornith-1.0 9B | Use when the workload needs coding, 262k context, and 9B parameters. | 2026-06 | coding262k context9B parameters | Current |
| Ornith-1.0 35B | Use when the workload needs coding, 262k context, and 35B parameters. | 2026-06 | coding262k context35B parameters | Current |
| Ornith-1.0 397B | Use when the workload needs coding, 262k context, and 397B parameters. | 2026-06 | coding262k context397B parameters | Current |
Release Timeline
1 release groupSpecifications(3 models)
| Model | Released | Context | Parameters | Reasoning | Fn Calling | Tool Use |
|---|---|---|---|---|---|---|
| Ornith-1.0 9B | 2026-06 | 262k | 9B | Yes | Yes | Yes |
| Ornith-1.0 35B | 2026-06 | 262k | 35B | Yes | Yes | Yes |
| Ornith-1.0 397B | 2026-06 | 262k | 397B | Yes | Yes | Yes |
Frequently Asked Questions
- What is Ornith 1.0 used for?
- Ornith 1.0 is used for coding, reasoning, and agent workflows and tool use. The family description and listed model capabilities point to those workloads as the best fit.
- How does Ornith 1.0 compare to Claude 3?
- Ornith 1.0 by DeepReinforce is strongest where you need coding, while Claude 3 by Anthropic is the closest related family to check for vision and multimodal work. Ornith 1.0 has 3 listed variants and reaches up to 262k context, while Claude 3 reaches up to 200k context, so compare the specs and pricing tables before choosing a production model.
- Which Ornith 1.0 model should I use?
- If price is the main constraint, use the pricing table first because Ornith 1.0 does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Ornith-1.0 9B with 262k context and reasoning, tool use, and function calling.