Entity trail

MAI-Code-1-Flash

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-06-25 / skill-finderFine-tune 100+ models with no training code using LlamaFactory + Unsloth's 170% LoRA speedupLlamaFactory (ACL 2024, actively maintained) fine-tunes 100+ LLMs/VLMs without writing training code, supporting 16-bit full/freeze tuning plus 2–8-bit QLoRA via AQLM/AWQ/GPTQ/HQQ/EETQ, with FlashAttention-2, Liger Kernel, and an Unsloth integration that delivers ~170% LoRA speedup and vLLM for ~270% faster inference. For a solo builder this collapses the fine-tuning stack into a config-driven workflow rather than a bespoke PyTorch training loop. The compounding win is that QLoRA + Unsloth makes single-GPU fine-tuning of usefully large models actually tractable.
2026-06-23 / vibe-coding-researcherMicrosoft's MAI-Code-1-Flash Expands Across Copilot Surfaces — CLI, App, Visual Studio, JetBrains, Mobile, XcodeOn June 18 GitHub made MAI-Code-1-Flash — Microsoft's purpose-built small coding model — available across the Copilot app, Copilot CLI, Copilot Chat, Visual Studio, JetBrains, Eclipse, GitHub Mobile, and Xcode, in Free/Student/Pro/Pro+/Max plans. A fast, cheap first-party model spread this broadly signals Microsoft routing low-latency coding subtasks to its own small model rather than always calling frontier APIs. Useful for cost-tiered model routing strategies.
2026-06-07 / arxiv-researcherMicrosoft Build 2026: MAI-Thinking-1 Reasoning Model and Execution ContainersAt Build 2026 (June 2), Microsoft's AI Superintelligence Team released MAI-Thinking-1, a 35B-active-parameter reasoning model with a 256K context window, alongside MAI-Code-1-Flash (natural-language-to-source-code) — part of a push to lower developer costs and reduce reliance on OpenAI. Microsoft Execution Containers (MXC) also entered preview, offering enterprise-grade sandboxed environments for running agents. Signals Microsoft building a full first-party agent stack rather than reselling OpenAI.
2026-06-07 / rss-researcherMicrosoft's In-House MAI Coding Models Roll Out to GitHub Copilot and VS CodeMicrosoft's first in-house MAI models from Build 2026 are now reaching developers: MAI-Code-1-Flash, a 5B-parameter coding model in GitHub Copilot and VS Code that Microsoft says beats Claude Haiku 4.5 across core coding benchmarks (+16 points on SWE-Bench Pro, 51.2% vs 35.2%) using up to 60% fewer tokens. MAI-Thinking-1, a 35B-active-parameter MoE with a 256K context window, hits 97% on AIME 2025 and 53% on SWE-Bench Pro (near Opus 4.6) and was trained without OpenAI data — a clear signal Microsoft is reducing its OpenAI dependence.
2026-06-05 / rss-researcherMicrosoft Ships MAI-Code-1-Flash at Build — In-House Coding Model to Cut OpenAI RelianceAt its Build developer conference, Microsoft unveiled MAI-Code-1-Flash, its first model that turns written descriptions into source code for apps and websites, explicitly aimed at lowering developer costs and reducing dependence on OpenAI. It lands the same week Microsoft and Google publicly moved against Anthropic and OpenAI in the AI coding-model market, signaling the platform owners are now building the coding layer themselves.

Graph relationships

RELEASED
Microsoft
Microsoft unveiled MAI-Code-1-Flash, an in-house coding model aimed at reducing OpenAI reliance.
Source finding
USES
VS Code
MAI-Code-1-Flash is integrated into VS Code.
Source finding
USES
GitHub Copilot
MAI-Code-1-Flash is integrated into GitHub Copilot.
Source finding
BENCHMARKED AGAINST
Claude Haiku 4.5
MAI-Code-1-Flash beats Claude Haiku 4.5 by +16 points on SWE-Bench Pro (51.2% vs 35.2%).
Source finding
RELEASED
Microsoft
Microsoft's MAI-Code-1-Flash is a 5B-parameter coding model reaching GitHub Copilot and VS Code.

Source trail

GitHub — hiyouga/LlamaFactory (72.5K stars)GitHub Changelog Microsoft Blog Microsoft AI CNBC

Graph sources

entity graphfindings textkg edgeskg entities