Official components · needs a real run0 successful reportsReviewed 2026-08-16

Run Qwen3 8B Q4 in LM Studio on a 16GB Apple Silicon Mac

A desktop-first local chat and API path using LM Studio, a Qwen3 8B Q4_K_M GGUF artifact, Metal acceleration, and an 8K baseline context.

Report result

Evidence status

LM Studio officially recommends 16GB RAM on Apple Silicon and supports GGUF models. The exact Qwen3 artifact and baseline remain unverified by LocalAIRun.

Requirements

  • Apple Silicon M1 or newer
  • macOS 13.4 or newer
  • 16GB unified memory recommended by LM Studio
  • At least 12GB free disk space

Stack

Desktop runtime
LM Studio · Current stable release

LM Studio supports Apple Silicon and GGUF models through llama.cpp; MLX is also available for matching artifacts.

Model artifact
Qwen3 8B Q4_K_M · GGUF Q4_K_M

Select a clearly identified Q4_K_M GGUF build and review its source repository before downloading.

Manual mode

Step-by-step execution

0 / 4 complete

Use the same safeguards as the Skill

Check the machine first, use only the linked official sources, confirm large downloads and file locations, and do not bypass macOS security. Mark a step complete only after its check passes.

  1. 1. Check LM Studio requirements

    Official instructions →
    • •Confirm Apple Silicon, macOS 13.4 or newer, and 16GB unified memory.
    • •Close memory-heavy apps before loading the model.
    Check: The Mac meets LM Studio's published requirements.

  2. 2. Install LM Studio

    • •Download and install LM Studio from its official site.
    • •Open the app and verify that an Apple Metal runtime is available.
    Check: LM Studio launches and recognizes the Apple Silicon runtime.

  3. 3. Download and load Qwen3 8B

    • •Search for Qwen3 8B and select a Q4_K_M GGUF artifact from a traceable source.
    • •Load it with an 8K context for the baseline run.
    Check: The model loads and memory pressure remains usable.

  4. 4. Test chat and local API

    • •Run the fixed prompt in Chat.
    • •Start LM Studio's local server and repeat the prompt through its OpenAI-compatible endpoint.
    • •Record versions and result timing before submitting a report.
    Check: Chat and API calls complete or the exact failure is recorded.

Baseline settings

context length
8192
temperature
0.2
gpu offload
auto

Comparable test

Prompt

Write a small TypeScript function that groups hardware variants by product ID, then explain its time complexity.

Seed: 20260816

Did it run on your hardware?

Success and failure reports both improve this recipe's evidence level.

Submit test result

Sources

LM Studio system requirements →LM Studio basics →