# Can I run Sarvam-105B on iPhone 16?

> **No, not enough memory.** Needs ~67.5 GB even at Q4_K_M, but only ~4.5 GB is usable.

Last validated: 2026-06-15.

## Memory math
- Sarvam-105B: 105B (MoE, 10.3B active); Q4_K_M weighs 64.2 GB on disk.
- At Q4_K_M with a 4k context it needs ~67.5 GB (weights + KV cache + ~0.8 GB runtime overhead, ±15% with context length).
- iPhone 16: 8 GB unified, ~4.5 GB usable for model weights.
- Headroom: ~-63 GB.

## How to run
Recommended quant: n/a. Best tool on iOS: Apple Foundation Models.

## Run Sarvam-105B on other devices
- [Apple M4 Max (128GB)](https://localmodel.run/can-i-run/sarvam-105b/apple-m4-max-128gb)
- [Apple M5 Max (128GB)](https://localmodel.run/can-i-run/sarvam-105b/apple-m5-max-128gb)
- [Apple M3 Ultra (256GB)](https://localmodel.run/can-i-run/sarvam-105b/apple-m3-ultra-256gb)

## Other models that run on iPhone 16
- [Gemma 3 4B (4B)](https://localmodel.run/can-i-run/gemma-3-4b/iphone-16)
- [Qwen3 4B (4B)](https://localmodel.run/can-i-run/qwen3-4b/iphone-16)
- [Phi-3.5-mini 3.8B (3.82B)](https://localmodel.run/can-i-run/phi-3.5-mini/iphone-16)
- [Phi-4-mini 3.8B (3.8B)](https://localmodel.run/can-i-run/phi-4-mini-3.8b/iphone-16)
- [Qwen2.5-VL 3B (3.75B)](https://localmodel.run/can-i-run/qwen2.5-vl-3b/iphone-16)
- [Qwen2.5 3B (3.09B)](https://localmodel.run/can-i-run/qwen2.5-3b/iphone-16)

Estimate, not a guarantee. Sources: https://huggingface.co/sarvamai/sarvam-105b, https://huggingface.co/sarvamai/sarvam-105b-gguf, https://www.sarvam.ai/blogs/sarvam-30b-105b.
More: https://localmodel.run/can-i-run/sarvam-105b/iphone-16