The LLMCheck index estimates DeepSeek V4 Flash Vision-Exp at ~32 tok/s on an M3 Ultra at Q4_K_M, a figure from its published bandwidth model, not a measured run. It needs about 120 GB of unified memory. It is DeepSeek's 305B mixture-of-experts model (13B active), released under the MIT licence.
Want to run DeepSeek V4 Flash Vision-Exp 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~32 tok/s |
| Provenance | estimate · how estimates are made |
| Quantization | Q4_K_M |
| Engine | llama.cpp |
| Figure dated | 2026-09 |
| Minimum RAM | 120 GB |
| Parameters | 305B 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 | llama.cpp | 256 GB | — | LLMCheck estimator | — |
Ollama — the index has not verified an Ollama tag for DeepSeek V4 Flash Vision-Exp, 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 → · 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 first V4 vision model (weights 2026-08-31, MIT): V4 Flash's 13B-active decode path with vision modules added, 305B total, explicitly experimental. Text capability inherited from V4 Flash; no independent vision benchmark verified. 2-bit GGUF is 97 GB and fits a 128 GB Mac; image input needs llama.cpp b10766 or later. ~39 tok/s estimated on M5 Max, anchored to V4 Flash's community run.
MIT licence · 305B MoE · 1000K context · LLMCheck capability 47/50 · see its leaderboard rank →
DeepSeek V4 Flash Vision-Exp needs about 120 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 DeepSeek V4 Flash Vision-Exp stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard