uran1um1/tenorio-0.6b

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 20, 2026License:mitArchitecture:Transformer0.0K Open Weights Featherless Exclusive Cold

Tenorio 0.6b by uran1um1 is an 0.8 billion parameter, 32768-token context length model, fine-tuned from Alibaba's Qwen3 0.6b. It is specifically optimized for enhanced proficiency in modern Spanish, focusing on improved formatting, grammatical correctness, and natural language responses. This model is designed for applications requiring high-quality Spanish output without complex reasoning capabilities.

Loading preview...

Tenorio 0.6b: Spanish Language Optimized Model

Tenorio 0.6b is a compact, 0.8 billion parameter model developed by uran1um1, fine-tuned from Alibaba's Qwen3 0.6b. Its primary focus is to deliver superior performance in the modern Spanish language, making it suitable for applications where high-quality Spanish output is crucial.

Key Capabilities and Training:

  • Enhanced Spanish Proficiency: The model has undergone continued pretraining on over 100 public domain Spanish books and instruction tuning with 7077 Spanish Q&A pairs.
  • Improved Output Quality: It exhibits better context awareness, reduced hallucination, and more concise and natural answers when responding in Spanish.
  • Reasoning Destruction: Deliberately designed to minimize reasoning abilities, which helps in reducing hallucinations and preventing infinite loops, leading to more direct and accurate Spanish responses.

Benchmarks:

While developed as a hobby project, custom benchmarks indicate its effectiveness in Spanish language quality. Using a GPT-OSS 20b model to judge grammar and language quality, Tenorio 0.6b showed improved performance over its base model in various categories:

  • Creative writing: 40/80
  • General STEM: 42/80
  • Coding & CS: 45/80

These scores reflect its ability to produce grammatically sound and high-quality Spanish text, rather than content accuracy or reasoning.

Good For:

  • Applications requiring precise and natural Spanish language generation.
  • Use cases where complex reasoning is not needed, prioritizing clear and grammatically correct Spanish communication.
  • Edge or tiny model deployments due to its compact size.