Skip to content
AI Model Radar

FastContext 1.0 4B SFT

microsoft/FastContext-1.0-4B-SFT
Open Source · chat · open-weights
Unavailable Alert me on changes
Not currently available. This model's weights can't currently be fetched from Hugging Face, so it isn't usable right now. We re-check every few minutes and will alert subscribers the moment it returns.
Context
262K
Max output
—
Weights
Unavailable
API $/1M
—
Modalities
text
Released
14 Jun 2026
License: mit · microsoft/FastContext-1.0-4B-SFT
Download image Share on X Share on LinkedIn
AI summary
● machine-written

Microsoft releases FastContext 1.0 4B SFT for repository exploration

FastContext 1.0 4B SFT is a 4 billion parameter model from Microsoft designed as a repository-exploration subagent for coding agents. It performs parallel tool calls to extract focused code context, reducing main-agent token consumption by up to 60.3% while improving end-to-end accuracy by up to 5.5% on benchmarks like SWE-bench Pro.

What's new
  • 4B parameter model optimized for efficient repository exploration in coding tasks
  • Supports 262K token context length for large codebase navigation
  • Achieves up to 60.3% token savings on main-agent queries
  • Up to 5.5% accuracy improvement on SWE-bench Pro benchmark
  • Available under MIT license as open-weights model
Best for
Reducing token consumption in LLM-powered coding agentsFast and precise code repository explorationImproving accuracy of main coding agents with focused context
Sources

Source: https://huggingface.co/microsoft/FastContext-1.0-4B-SFT