DeepSeek-V4-Flash VRAM requirements
DeepSeek-V4-Flash's smallest measured build is unsloth's UD-IQ1_S, 1-bit. Three doors are real this time: every tracked build fits a single 192 GB B200 (1× at the smallest, 1× at NVFP4); a 512GB Mac Studio runs all of them; and a reported community path runs the 1-bit build on a 24GB card with 96GB of system RAM (tensorrigs, linked — offload, not VRAM). The official-precision FP8 build totals 156.2 GiB.
Generation note: this page answers for the V4.0 generation — the 0731 weights are still the current DeepSeek-V4-Flash release on Hugging Face and the files most holders run. Since the V4.1-Flash release of September 10, 2026 the deepseek-v4-flash API name routes to the newer V4.1-Flash generation (primary name: deepseek-flash). Which version is this page? A V4.1-Flash page links from here when tracked.
Updated · data status: MEASURED WEIGHTS — totals = measured weights + reported KV (source-published 1M figure, linear-rescaled down to 32K — over-counts at short context (the source's own note), direction: scaled down, not measured at 32K) + measured overhead; every weight is a pinned file-size sum (unsloth / ggml-org / bartowski / mlx-community / nvidia / deepseek-ai)
Rent it: minimal cloud configs (the smallest door in)
| Quantization | Total VRAM | Minimal single-node config | Verdict |
|---|---|---|---|
| UD-IQ1_S (measured) | 77.7 GiB | 1× B200 / 1× MI300X | EXTREME · 1-bit trade-off |
| UD-IQ2_XXS (measured) | 85.4 GiB | 1× B200 / 1× MI300X | EXTREME · 2-bit dynamic |
| UD-IQ4_NL (measured) | 128.1 GiB | 1× B200 / 1× MI300X | single-node door · cheapest Q4-named build (classify ratio 5.34×) |
| UD-Q4_K_XL (measured) | 145.2 GiB | 1× B200 / 1× MI300X | single-node door · the Q4_K_M-equivalent |
| FP8 (official deepseek-ai) | 156.2 GiB | 1× B200 / 1× MI300X | the official-precision door |
| NVFP4 (nvidia) | 164.3 GiB | 1× B200 / 1× MI300X | single-node door · NVIDIA format |
Card counts are plain division (rounded up) against published per-GPU memory (B200 192GB, MI300X 192GB) on measured totals — weights + a reported KV line + measured overhead. The KV line is arithmetic from the source's published 1M figure (reported, linked; the source's own note says it over-counts at short context), so totals are labeled provisional, not measured — at 3× of a 24GB card no consumer row is anywhere near the line, and that verdict doesn't move. Every door below is a single node.
Appliance / unified-memory form factors
| Appliance | Unified memory | Best fit | Verdict |
|---|---|---|---|
| Mac Studio M3 Ultra | 512 GB | every build through NVFP4 (164.3 GiB) fits — 6.24× of smallest (usable), a real door | FITS · 1-bit through NVFP4 |
| Mac Studio M4 Max | 128 GB | builds through Q2_K (110.1 GiB) fit — 1.56× of smallest | FITS · 1–2-bit + Q2 doors |
| NVIDIA DGX Spark | 128 GB | builds through Q2_K (110.1 GiB) fit — 1.56× of smallest | FITS · 1–2-bit + Q2 doors |
The appliance column is a real exit and a wide one: the 512GB Mac Studio (480 GiB usable) holds every tracked build through 164.3 GiB NVFP4 — 6.24× of the measured 76.9 GiB smallest file — and the 128GB boxes (120 GiB usable, 1.56×) clear the builds through the 110.1 GiB Q2_K door. Ratios compare usable memory to build file size, both figures in GiB (binary) — the division is computed and checked in a single unit end to end, so decimal-GB rereadings of these numbers do not apply; unified-memory bandwidth trails datacenter GPUs by an order of magnitude, so these are "it runs" verdicts, not "it runs fast".
Can't run it? The distill / step-down ladder
No official DeepSeek-V4-Flash distill is tracked in our sources yet. Architecture note: attention is a hybrid of CSA and HCA — Compressed Sparse + Heavily Compressed, interleaved across layers — the successor of the V3-era MLA line (pinned sources below). A V4-Flash distill or step-down sibling would instantly become the family's consumer entry — a native small model beats a 1-bit 284B every time. When official distills or smaller siblings ship, each gets its own VRAM page and this section links the full ladder; until then, the smallest real door is the pinned 1-bit build in the rent table above.
Step-down options that exist: DeepSeek's own R1-Distill line from the V3 generation — R1 reasoning distilled into Qwen (1.5B/7B/14B/32B) and Llama (8B/70B) bodies, real files on Hugging Face (provenance below) — is the runnable DeepSeek-family answer while V4-Flash has no distill of its own; the V3 base generation itself is not yet tracked on this site.
Private deployment (enterprises, labs, providers)
A separate query family from everything above: vLLM / SGLang tensor-parallel reference configs for the CSA/HCA stack, compliance and air-gapped deployments, TCO modeling against the API. The buyers here are not price-shopping: regulated workloads, data-residency rules, and air-gapped labs make self-hosting a procurement requirement, not an optimization. And the capacity fact flips the usual deploy math: a single B200 node already holds every tracked build, so the question is governance, not RAM.
Or just use the API
Honest take: DeepSeek-V4-Flash is a 284.0 B MoE with 13.0 B active parameters — the shape where hosted APIs are cheapest to run and hardest to beat on cost: 13B-active inference is exactly what API fleets optimize. Self-hosting makes sense when private deployment is the requirement, not to save money. Prices are linked, not pinned.
Where every number comes from (every figure carries its tag)
| Tag | Slot | What it is | What the verdict can claim |
|---|---|---|---|
| measured | Build file sizes | Community + official quant files, sizes summed at pinned revisions (unsloth / ggml-org / bartowski 0731 GGUF, mlx-community 2.4bit, nvidia NVFP4, deepseek-ai FP8 — linked below). | Determinate at each door — a platform above or below a measured file size is a fact, not an estimate. |
| provisional | Totals (weights + KV + overhead) | Simple addition: measured weights + reported KV (arithmetic from the source's published 1M bf16 total, linear-rescaled to 32K; the source's own note: it over-counts at short context, so read it as a short-ctx ceiling, not a measurement) + a measured overhead anchor (llama.cpp discussion #9936). | The KV line is a rounding-level term next to the weights — totals sit within about a GiB of the weights alone; at 3× of a 24GB card, no door verdict moves on this line. |
| provisional | Cloud card counts | Simple division against published per-GPU memory (B200 192GB, MI300X 192GB — linked below). | Card counts follow arithmetically once totals are set. |
| reported | Offload path (24GB card + 96GB RAM) | Single community source — tensorrigs hardware guide (accessed 2026-09-08), linked, not re-measured by us. | One self-declared "cheapest entry for 300B-class" route: weights live in system RAM, the GPU takes attention + active experts. A reported route, not a verified verdict. |
| reported | API prices | Referenced from OpenRouter, linked, not re-verified by us. | Tracks the linked page — prices move, we don't pin them. |
FAQ
Can DeepSeek-V4-Flash run on an RTX 4090?
Not in VRAM — the smallest measured build (unsloth UD-IQ1_S) totals 77.7 GiB, 3× the card's 24GB. A reported community path runs the 1-bit build on a 24GB card with 96GB of system RAM (tensorrigs, linked — offload, at a fraction of datacenter speed). The in-VRAM paths are a single B200, a 512GB Mac Studio, or the API.
Can a 96GB workstation card (RTX PRO 6000 Blackwell) run DeepSeek-V4-Flash?
At 32K context the smallest build's total (77.7 GiB) is 0.8× of the card's 96GB — it loads on paper. At the model's full 1M context the KV line grows to the source's published 1M figure (linked below), pushing the smallest build's total to within a few GiB of the card's entire memory — no room left for activations and system: practically no. We promise order-of magnitude, not decimal-precision, on this edge.
How many GPUs do you need to run DeepSeek-V4-Flash?
One, at every tracked build: even NVFP4 (164.3 GiB) fits a single 192GB B200 (card counts are plain division on measured totals). A 5× B200 node is a throughput choice, not a capacity requirement.
Can a Mac Studio run DeepSeek-V4-Flash?
Yes — and unusually widely. The 512GB M3 Ultra (480 GiB usable) runs every tracked build through the 164.3 GiB NVFP4; the 128GB boxes (M4 Max / DGX Spark, ~120 GiB usable) run the builds through the 110.1 GiB Q2_K door. Unified-memory bandwidth caps throughput, so treat these as "it runs" verdicts.
Is there a smaller version of DeepSeek-V4-Flash?
No official distill is tracked in our sources yet (attention is a CSA / HCA hybrid — see the distill section). When a V4-Flash distill or smaller sibling ships it becomes the practical consumer entry, and this page will link it. Until then there is no consumer-sized door: the smallest measured build is the pinned 1-bit in the rent table above.
Can I run DeepSeek-V4-Flash in Ollama?
Not locally from the official library — Ollama's official listing is cloud-only (the :cloud tags; Ollama hosts it, no local weights are offered). A community import exists (frob/deepseek-v4-flash-0731 — unsloth's 0731 GGUF re-listed; needs Ollama v0.31.1+). The well-trodden local doors skip Ollama entirely: llama.cpp mainline, LM Studio, and Unsloth Studio load the tracked builds directly; the tight-machine recipes in the community (1-bit build, KV-cache flags, layer offload) are llama.cpp-ecosystem recipes.
Which DeepSeek-V4-Flash am I looking at — V4.0 or V4.1?
This page answers for the V4.0 generation: every number here prices the 0731 weights, which are still the current DeepSeek-V4-Flash release on Hugging Face and the files most holders run. The API side has split: since the V4.1-Flash release of September 10, 2026, the deepseek-v4-flash API name is retired and temporarily routes to the newer V4.1-Flash generation, whose primary name is deepseek-flash (official notice in provenance) — the API name you call is no longer proof of which weights you get. For version-pinned answers ("the July build specifically"), use the DeepSeek-V4-Flash-0731 page; a V4.1-Flash page links from here when tracked.