SynastriaNetworks/U1-0817

VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 17, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

SynastriaNetworks/U1-0817 is a 9 billion parameter Uirapuru U1 language model, developed by SynastrIA Networks, focused on refining identity, reasoning, and agentic task performance. This hotfix release emphasizes improved Brazilian Portuguese fluency and consistent reasoning behavior by incorporating learned behavior directly rather than adaptive thinking. It is optimized for conversational AI, agentic systems, and applications requiring natural informal Brazilian Portuguese.

Loading preview...

Overview

SynastriaNetworks/U1-0817 is an incremental hotfix release of the Uirapuru U1 language model, developed by SynastrIA Networks. This 9 billion parameter model focuses on refining and enhancing the Uirapuru U1's core capabilities rather than introducing a completely new generation. A key change in this iteration is the retraining without adaptive thinking, instead embedding reasoning behavior directly into the model's learned patterns for more consistent performance. It also utilizes a higher LoRA rank to provide increased capacity for learning behavioral and capability-level changes.

Key Improvements & Capabilities

  • Stronger Identity: A more consistent Uirapuru identity and behavioral profile.
  • Improved Reasoning: Enhanced and more consistent reasoning patterns, with reasoning behavior directly incorporated into learned behavior.
  • New Capabilities: Additional skills learned during retraining.
  • Informal Brazilian Portuguese: More natural and fluent conversational Portuguese, including everyday language.
  • Agentic Task Performance: Improved ability to follow objectives, handle multi-step instructions, and perform task-oriented workflows.

Intended Use Cases

  • AI assistants and conversational applications.
  • Applications requiring strong Brazilian Portuguese fluency.
  • Agentic AI systems and task automation.
  • LLM experimentation, research, and prototyping.