Ilides/Spartan-1.2-2B-MLX-8bit

Spartan 1.2 2B (MLX 8-bit)

Fine-tuned Qwen3.5-2B by Ilides. Agentic coding focus with anti-loop behavior.

• Format: MLX 8-bit (Apple Silicon)
• Base: Spartan 2B → Qwen3.5-2B
• Use with: LM Studio, mlx-lm, Cortex

Load (mlx-lm)

pip install mlx-lm
python -m mlx_lm.generate --model Ilides/Spartan-1.2-2B-MLX-8bit --prompt "Hello"

LM Studio

Import this repo folder or load from local path. Use system prompt: workspace-relative paths, Write for files.

GGUF variant

See Ilides/Spartan-1.2-2B-GGUF for llama.cpp / Ollama.