wesleysimplicio/Simplicio-27B

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Oct 4, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

Simplicio 27B is a 27.36 billion parameter LoRA fine-tune of Qwen3.8-27B by Wesley Simplicio at SimpleTI, designed to generate code-change requests using SEARCH/REPLACE patches. This model specializes in providing surgical code edits within a five-phase development loop, focusing on modifying only the necessary lines rather than entire files. It supports a 32768-token context length and is optimized for developer workflows requiring precise, diff-based code modifications.

Loading preview...

Simplicio 27B: Code-Change Generation with SEARCH/REPLACE Patches

Simplicio 27B, developed by Wesley Simplicio at SimpleTI, is a LoRA fine-tune of the Qwen3.8-27B base model. It is specifically trained to respond to code-change requests by generating SEARCH/REPLACE patches, focusing on surgical edits rather than outputting entire modified files. This approach aims to streamline developer workflows by providing precise, line-level code modifications.

Key Capabilities

  • Diff-based Code Generation: Generates code changes in a SEARCH/REPLACE patch format, targeting only the lines that require modification.
  • Structured Development Workflow: Operates within a five-phase (<orient>, <plan>, <patch>, <validate>, <deliver>) Simplicio-Loop, guiding the code modification process.
  • Qwen3.8-27B Base: Built upon the Qwen3.8-27B architecture, featuring 27.36 billion parameters and a 32768-token context length.
  • Agent Integration: Provides a native Model Context Protocol (MCP) server with tools for applying surgical diffs, validating patch structures, and formatting prompts.
  • Diff Format Compatibility: Supports both Aider standard (7-char) and Simplicio compact (4-char) diff markers.

Good For

  • Automated Code Refactoring: Ideal for scenarios requiring automated, precise code modifications and refactoring tasks.
  • Developer Tool Integration: Suitable for integration into developer environments and agents that benefit from structured, diff-based code generation.
  • Code Review Automation: Can assist in generating proposed changes that are easy to review due to their granular nature.
  • Research into Code Generation: Useful for researchers exploring fine-tuning large language models for specific code manipulation tasks and structured outputs.

Finetunes
1 models