VSCode Continue.dev LMstudio models for Apple Silicon

aciidix 's Collections

updated Jan 21

If you are using Apple Silicon and VSCode and Continue.Dev and LMStudio then heres the recommended models. The yaml spacing you have to do again.

Upvote

Qwen/Qwen3-Coder-30B-A3B-Instruct

Text Generation • 31B • Updated Dec 3, 2025 • 2.49M • • 1.02k

Note - name: qwen3 coder 30b provider: lmstudio model: qwen/qwen3-coder-30b apiBase: http://localhost:1234/v1/
aciidix/instinct-mlx-4Bit

Text Generation • 1B • Updated Jan 20 • 22

Note - name: instinct provider: lmstudio model: instinct-mlx roles: - autocomplete defaultCompletionOptions: temperature: 0 stop: - <|im_end|> apiBase: http://localhost:1234/v1/
aciidix/FastApply-1.5B-v1.0-mlx-4Bit

Text Generation • 0.2B • Updated Jan 20 • 43

Note - name: fastapply provider: lmstudio model: fastapply-1.5b-v1.0-mlx roles: - apply apiBase: http://localhost:1234/v1/
mlx-community/Qwen3-Embedding-0.6B-4bit-DWQ

Text Generation • Updated Jun 6, 2025 • 91k • 8

Note - name: qwen3 embedding provider: lmstudio model: qwen3-embedding-0.6b-dwq roles: - embed apiBase: http://localhost:1234/v/
galaxycore/Qwen3-Reranker-8B-MLX-4bit

Text Generation • 1B • Updated Sep 25, 2025 • 191 • 1

Note - name: Qwen3 reranker provider: lmstudio model: qwen3-reranker-8b-mlx apiBase: http://localhost:1234/v1/
wobondar/instinct-mlx-4Bit

Text Generation • 1B • Updated Sep 24, 2025 • 28

Note - name: instinct provider: lmstudio model: instinct-mlx roles: - autocomplete defaultCompletionOptions: temperature: 0 stop: - <|im_end|> apiBase: http://localhost:1234/v1/
mistralai/Codestral-22B-v0.1

22B • Updated Jul 24, 2025 • 8.25k • 1.33k

Note models: - name: Codestral provider: mistral model: codestral-latest apiBase: http://localhost:1234/v1/
Qwen/Qwen3-Embedding-0.6B

Feature Extraction • 0.6B • Updated 1 day ago • 5.96M • • 990

Note - name: qwen3 embedding provider: lmstudio model: qwen3-embedding-0.6b-dwq roles: - embed apiBase: http://localhost:1234/v/
lmstudio-community/Qwen2.5-Coder-1.5B-Instruct-MLX-8bit

Text Generation • 0.4B • Updated Nov 13, 2024 • 2.78k • 1

Note - name: qwen 2.5 coder 1.5b Autocomplete Model provider: lmstudio model: qwen2.5-coder-1.5b-instruct-mlx roles: - apply - autocomplete - chat - edit apiBase: http://localhost:1234/v1/

Upvote