devehz/MagenticBrain-GGUF overview
MagenticBrain GGUF GGUF quantization of microsoft/MagenticBrain https://huggingface.co/microsoft/MagenticBrain 14B for use with llama.cpp and compatible runner…
Runs locally from ~4.00 GB disk (8 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| MB-BF16.gguf | GGUF | BF16 | 27.51 GB | Download |
| MB-IQ2_XXS.gguf | GGUF | IQ2_XXS | 4.00 GB | Download |
| MB-IQ3_M.gguf | GGUF | IQ3_M | 6.41 GB | Download |
| MB-IQ3_XXS.gguf | GGUF | IQ3_XXS | 5.53 GB | Download |
| MB-IQ4_XS.gguf | GGUF | IQ4_XS | 7.55 GB | Download |
| MB-Q2_K.gguf | GGUF | Q2_K | 5.36 GB | Download |
| MB-Q3_K_M.gguf | GGUF | Q3_K_M | 6.82 GB | Download |
| MB-Q4_K_M.gguf | GGUF | Q4_K_M | 8.38 GB | Download |
| MB-Q4_K_S.gguf | GGUF | Q4_K_S | 7.98 GB | Download |
| MB-Q5_K_M.gguf | GGUF | Q5_K_M | 9.79 GB | Download |
| MB-Q6_K.gguf | GGUF | Q6_K | 11.29 GB | Download |
| MB-Q8_0.gguf | GGUF | Q8_0 | 14.62 GB | Download |
Model Details
| Model ID | devehz/MagenticBrain-GGUF |
|---|---|
| Author | devehz |
| Pipeline | text-generation |
| License | mit |
| Base model | microsoft/MagenticBrain |
| Last modified | 2026-08-13T02:59:52.000Z |
Model README
---
license: mit
base_model: microsoft/MagenticBrain
library_name: llama.cpp
pipeline_tag: text-generation
tags:
- gguf
- llama.cpp
- quantized
- qwen3
- agents
---
MagenticBrain-GGUF
GGUF quantization of microsoft/MagenticBrain (14B) for use with llama.cpp and compatible runners (Lemonade, LM Studio, Ollama, etc.).
Available quants
| File | Quant | Size | Notes |
|---|---|---|---|
| MB-Q6_K.gguf | Q6_K | ~11.5 GB | Recommended — near-lossless; preserves the model's structured tool-call / JSON fidelity |
| MB-Q4_K_M.gguf | Q4_K_M | ~9 GB | Fallback for tight VRAM budgets |
Converted with convert_hf_to_gguf.py (BF16 intermediate) and quantized with llama-quantize from llama.cpp. Standard k-quants, no imatrix calibration.
About the model
MagenticBrain is Microsoft Research AI Frontiers' 14B orchestration model, SFT'd from Qwen3-14B on agentic data — function calling (APIGen-MT, ToolACE, xLAM), 250+ synthetic MCP environments, file-system and terminal trajectories, and sub-agent delegation traces (handoffs to Fara1.5-9B) — followed by an RL stage on terminal tasks. It plans multi-step tasks, calls declared tools via structured JSON (never inventing new ones), coordinates sub-agents, and follows a submit-to-terminate protocol.
This is an orchestration-first model, not a general-purpose chat model. It is co-designed with, and most thoroughly evaluated in, Magentic-Lite (Magentic-UI v0.2).
Recommended settings
- Context: up to 32K
- Thinking: disabled by default (
enable_thinking: false) — keep it off; verbose reasoning degrades long agentic trajectories - Tool calling: schemas are injected by the harness; the model expects to select only from declared tools
Usage
llama.cpp:
llama-server -m MB-Q6_K.gguf -ngl 99 -c 32768Lemonade / LM Studio / Ollama: register the GGUF file directly or pull this repo through the app's model management.
License & credit
MIT — same as the base model. All credit to Microsoft Research AI Frontiers. This repo is a community quantization and is not affiliated with or endorsed by Microsoft. See the base model card for full details.
Run devehz/MagenticBrain-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models