The LLMCheck index estimates MiniMax-M3 at ~42 tok/s on an M5 Ultra at Q4_K_M, scaled from its community-measured M3 Ultra run with the published bandwidth model, not a measured run on this chip. It needs about 162 GB of unified memory. It is MiniMax's 428B mixture-of-experts model (23B active), released under the MiniMax Community licence.
Want to run MiniMax-M3 faster — or step up to bigger models? Find your Mac →This is an LLMCheck index estimate from its published two-term bandwidth model, not a measured run; the index labels it so everywhere it appears. Own this Mac? Submit a real run →
| Metric | Value |
|---|---|
| Tokens per second | estimated~42 tok/s |
| Provenance | estimate · how estimates are made |
| Quantization | Q4_K_M |
| Engine | MLX |
| Figure dated | 2026-10 |
| Minimum RAM | 162 GB |
| Parameters | 428B MoE (23B active) |
| Chip bandwidth | 1228 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 (162 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 |
|---|---|---|---|---|---|---|
| estimated60 tok/s | IQ3_XXS | llama.cpp | 256 GB | — | LLMCheck estimator | — |
| estimated42 tok/s | Q4_K_M | MLX | 512 GB | — | LLMCheck estimator | — |
Ollama — the index has not verified an Ollama tag for MiniMax-M3, so it doesn’t print one (a guessed tag fails with “pull model manifest: file does not exist”). Search the Ollama library →
MLX — pick a build from Hugging Face’s MLX listings, then run it with its repo id:
Downloads: Hugging Face repo → · model card →
Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.
| Chip | Speed | Quant | Engine | Bandwidth |
|---|---|---|---|---|
| M3 Ultra | community29 tok/s | 4-bit | MLX | 819 GB/s |
MiniMax's 428B MoE (23B active, Jun 2026) with sparse attention, image and video input and a 1M window (512K guaranteed); SWE-bench Verified 80.5 on a Claude Code scaffold. The MiniMax Community License is non-commercial by default: commercial use needs a 'Built with MiniMax M3' credit and notice, plus written approval above $20M revenue. 3-bit GGUFs (~160 GB) fit a 192–256 GB Mac Studio and 4-bit needs 512 GB; a community oMLX run measured 28.9 tok/s at 4-bit on an M3 Ultra (1k context), and mlx-lm 0.32.0+ is text-only.
MiniMax Community licence · 428B MoE · 1000K context · LLMCheck capability 48/50 · see its leaderboard rank →
MiniMax-M3 needs about 162 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 MiniMax-M3 stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard