Can I run gpt-oss-120b on a M5 Ultra 96GB Mac?
Computed with the open FitLLM engine — accurate per-layer KV-cache modeling, not a naive estimate. Updated 2026-10-19.
Conditions: 8-bit · KV cache F16 · 32K context · macOS reserve (8.0 GB).
Memory breakdown (8-bit, F16 KV, 32K context)
| Model weights | 109 GB |
| KV cache | 1.1 GB |
| Runtime + macOS | 22.2 GB |
| Total used | 132 / 96 GB |
| Short by | 36.1 GB |
Context at 8-bit: does not fit at any context. Unified memory is shared with macOS and your apps — the recommended max leaves 20% of it free.
Every quantization on M5 Ultra 96GB
| Quant | Weights | Recommended max ctx | Hard limit | Used @32K |
|---|---|---|---|---|
| 4bit | 54.4 GB | ~113K | ~128K | ✅ 71.2 / 96 GB |
| 8bit | 109 GB | won't fit | ❌ 132 / 96 GB | |
| 16bit | 218 GB | won't fit | ❌ 254 / 96 GB | |
Recommended max = the longest context that still keeps the 20% headroom macOS needs · hard limit = the longest that fits at all (KV cache F16).
Lower quants free memory at some output-quality cost — 4-bit is the common sweet spot for local use.
▶ Open the interactive calculator (this exact setup)Embed this verdict
Live badge for your README or model card — recomputed by the engine at exactly these conditions, never stale:
[](https://fitllm.run/can-i-run/gpt-oss-120b-on-m5-ultra-96gb)
Or from your terminal (exit 0/1 — works as a pre-download guard):
npx fitllm "gpt-oss-120b" --mac 96 --quant 8 --ctx 32768 --kv 16Why most calculators get this wrong
gpt-oss-120b interleaves sliding-window (local) and global attention 5:1. The local layers cap their KV cache at the 128-token window, and the global layers use a different head shape (head_dim 64 vs 64). A naive "all layers × full context × one head_dim" formula over-counts KV cache by several times.
Other options
same Mac Models that fit in 96GB: GLM-4.7-Flash, gpt-oss-20b, Qwen 3.6 27B, Qwen 3.6 35B-A3B, Qwen-AgentWorld-35B-A3B, Gemma 4 e2b, Gemma 4 e4b, Gemma 4 12b, Gemma 4 26b A4B, Gemma 4 31b, Llama-3.2-3B-Instruct, Llama-3.1-8B-Instruct, MiniCPM5-1B, Qwen3-0.6B, Qwen3-1.7B, Llama-3.2-1B-Instruct, Gemma-3-1B-it, Qwen 3.8 27B, Laguna XS 2.1, Spark-X2.5-4B, Granite-4.2-30B, MiniCPM5-2B, Nex-N2.5-mini.
Reproduce it
Open math: fitllm-engine (MIT), from official config.json.
All numbers are computed by the open-source fitllm-engine (MIT) from official model config.json values — reproduce or audit them yourself. Estimates; real usage varies with runtime (llama.cpp / MLX / Ollama), driver and display. Found a mismatch? Report it. · FitLLM home