The LLMCheck index estimates DeepSeek V4 Flash at ~32 tok/s on an M3 Ultra at Q4_K_M, scaled from its community-measured M5 Max run with the published bandwidth model, not a measured run on this chip. It needs about 120 GB of unified memory. It is DeepSeek's 284B mixture-of-experts model (13B active), released under the MIT licence.
Want to run DeepSeek V4 Flash 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. The bandwidth formula’s typical error for large MoEs is ±34% (mostly below the measured figure), against 70 measured runs of 18 models (how it is computed). Own this Mac? Submit a real run →
| Metric | Value |
|---|---|
| Tokens per second | estimated~32 tok/s |
| Provenance | estimate · how estimates are made |
| Quantization | Q4_K_M |
| Engine | MLX |
| Figure dated | 2026-08 |
| Minimum RAM | 120 GB |
| Parameters | 284B MoE (13B active) |
| Chip bandwidth | 819 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 (120 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 |
|---|---|---|---|---|---|---|
| estimated32 tok/s | Q4_K_M | MLX | 256 GB | — | LLMCheck estimator | — |
Ollama — the index has not verified an Ollama tag for DeepSeek V4 Flash, 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: Search the Ollama library → · Hugging Face model card →
Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.
DeepSeek's 284B MoE model, ~39 tok/s on M5 Max. MIT license, LLMCheck capability 47/50.
MIT licence · 284B MoE · 1000K context · LLMCheck capability 47/50 · see its leaderboard rank →
Estimated, not measured: the index’s minimum for DeepSeek V4 Flash (120 GB — the weights at the quant that fits, plus ~2 GB) plus the KV cache for that many tokens, computed from its config.json. DeepSeek V4 compressed attention (repo inference/model.py): 21 layers keep 1/4-compressed latents (+ indexer keys), 20 keep 1/128, every layer a 128-token raw window; one 512-d latent per position, K and V shared. A 16-bit KV cache is the default in llama.cpp, Ollama, LM Studio and mlx-lm; a q8 cache halves the KV part. Coding agents hold 64K tokens or more. Runtimes also add a buffer that grows with context, so leave a few GB spare. macOS lets the GPU use only part of a Mac’s memory by default (figures for an M3 Ultra Mac on macOS 15, the lower default; macOS 26 allows more): “fits” means within the GPU memory a Mac of that tier can use with the limit raised, a tier tagged raise GPU limit fits only after sudo sysctl iogpu.wired_limit_mb raises it (the tooltip has the exact command), and the last column is the smallest Mac memory size whose default limit holds it (the GPU memory limit).
| Context | KV cache (16-bit) | RAM needed | RAM, q8 KV | Smallest tier that fits | Within the default GPU limit |
|---|---|---|---|---|---|
| 8K | ~0.1 GB | ~120.0 GB | ~120.0 GB | 192–256 GB Mac | |
| 32K | ~0.2 GB | ~120.2 GB | ~120.1 GB | 192–256 GB Mac | |
| 64K (coding-agent context) | ~0.5 GB | ~120.4 GB | ~120.2 GB | 192–256 GB Mac | |
| 128K | ~0.9 GB | ~120.9 GB | ~120.4 GB | 192–256 GB Mac |
DeepSeek V4 Flash needs about 120 GB of unified memory. These Apple Silicon Macs have enough to run it. Apple builds the higher-memory configurations to order; Amazon does not stock them, so the used option links Amazon Renewed listings.
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 DeepSeek V4 Flash stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard