Nemotron 3.5 Lightning runs at 72 tok/s (72.3 as recorded) on a 48 GB M5 Pro, a community run reported by github.com (LM Studio, Q4_K_M, 512-token coding answers, median of 5 at T 0.6). It needs about 18 GB of unified memory. It is NVIDIA's 30B mixture-of-experts model (3B active), released under the OpenMDW licence.
Want to run Nemotron 3.5 Lightning faster — or step up to bigger models? Find your Mac →This is a community run: a real run, reported by its source and linked below. The index records it as given and never adjusts it.
| Metric | Value |
|---|---|
| Tokens per second | community72 tok/s |
| Provenance | community run · github.com |
| Quantization | Q4_K_M |
| Engine | LM Studio |
| Context | 512-token coding answers, median of 5 at T 0.6 |
| Figure dated | 2026-10-19 |
| Minimum RAM | 18 GB |
| Parameters | 30B MoE (3B active) |
| Chip bandwidth | 307 GB/s |
Each row as the index holds it, with its source. “Mac RAM” is the memory of the Mac a run used, not the model’s requirement (18 GB). The headline figure follows the reference rule: the lower-median measured plain-decode run at 5 bits or fewer (a run whose quant was not stated counts only when no stated-quant run qualifies), else the estimate; a run marked “not used to anchor estimates” is listed here but never becomes the headline.
| Speed | Quant | Engine | Mac RAM | Context | Source | Note |
|---|---|---|---|---|---|---|
| community93.73 tok/s | 4-bit | MLX | 48 GB | 512-token coding answers, median of 5 at T 0.6 | github.com | MacBook Pro, 20-core GPU. mlx-lm 0.32.2 (MLX 0.32.4), mlx-community 4-bit, no MTP path; greedy 96.76; no contention. Run by the Kapitsa engine's developers with their own harness. |
| community72.3 tok/s | Q4_K_M | LM Studio | 48 GB | 512-token coding answers, median of 5 at T 0.6 | github.com | MacBook Pro, 20-core GPU. LM Studio llama.cpp runtime 2.60.0 (b11698), lmstudio-community Q4_K_M (no MTP head); greedy 72.2. Run by the Kapitsa engine's developers with their own harness. |
Ollama — install it (brew install ollama or the app from ollama.com), then pull the model by its tag, checked on ollama.com on 2026-10-24:
MLX — pick a build from Hugging Face’s MLX listings, then run it with its repo id:
Downloads: Search the Ollama library → · model card →
The 72 tok/s figure above came from LM Studio at Q4_K_M; other runtimes and quants run at different speeds.
Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.
NVIDIA's OpenMDW 30B MoE with 3B active and 1M context, built for throughput. It fits a 24 GB Mac at 4-bit; Liquid AI's published baseline lists 114 tok/s on an M5 Max (llama.cpp Q4_K_M), community runs on an M5 Max measured 116.8 (llama.cpp Q4_0) to 164.7 tok/s (oMLX 4-bit), 93.7 on an M5 Pro (mlx-lm 4-bit) and ~45 on a Mac mini M6.
OpenMDW licence · 30B MoE · 1000K context · LLMCheck capability 36/50 · see its leaderboard rank →
Nemotron 3.5 Lightning needs about 18 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep the LLMCheck index free.
See how Nemotron 3.5 Lightning stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard