text model · Sarvam · macOS
S Can I run Sarvam-105B on Apple M4 Pro (24GB)?
No. Sarvam-105B needs ~67.5 GB even at Q4_K_M, but Apple M4 Pro (24GB) only has ~16 GB usable.
Needs ~67.5 GB even at Q4_K_M, but only ~16 GB is usable.
- Q4_K_M needed
- ~67.5 GB
- Usable on device
- ~16 GB
- Device memory
- 24 GB
How to run it
On macOS use LM Studio (Polished GUI, ships MLX on Apple Silicon, one-click model downloads.). vLLM is NOT a Mac tool, it is a CUDA/Linux serving engine. Unified memory is not a fixed VRAM slice; ~70% is usable for weights.
- Parameters
- 105B (MoE, 10.3B active)
- Q4_K_M size
- 64.2 GB
- Context
- 128k
- Memory
- 24 GB unified
- Usable for weights
- ~16 GB
- Best runtime
- Ollama (MLX backend, preview) / MLX direct
What you can run instead
Run Sarvam-105B on other hardware
FAQ
Can Apple M4 Pro (24GB) run Sarvam-105B?
No. Sarvam-105B needs ~67.5 GB even at Q4_K_M, but Apple M4 Pro (24GB) only has ~16 GB usable.
How much memory does Sarvam-105B need?
Apple M4 Pro (24GB) does not have enough memory. At Q4_K_M the weights are ~64.2 GB; with KV cache and runtime overhead, budget ~67.5 GB at a 4k context. It is a Mixture-of-Experts model (105B total / 10.3B active), so all experts must stay in memory; memory tracks total params, not active params.
What is the best tool to run Sarvam-105B on macOS?
LM Studio for a simple setup; mlx-lm for the most speed. vLLM is NOT a Mac tool, it is a CUDA/Linux serving engine. Unified memory is not a fixed VRAM slice; ~70% is usable for weights.
Sources
Memory figures are estimates. See methodology.