DeepSeek-V4 Flash

DeepSeek describes the core MoE as 284B total with 13B active; the pinned weight archive contains about 304B tensors including auxiliary weights. Its single 512-dimensional KV head keeps long-context cache growth comparatively small.

DeepSeekMixture of expertscodingreasoning4 machines can run it

Total parameters

304.18B

Official spec sets memory need

Active parameters

13B

Official spec sets decode speed

Context

Official spec tokens

4-bit weights

166 GB

Calculated before KV cache

This is a sparse mixture of experts. All 304.18B of weights must sit in memory, but only 13B are read per generated token — so it needs the memory of a 304.18B model and generates at roughly the speed of a 13B one. That is why large unified-memory machines suit it and fast 32 GB GPUs do not.

Model specification

Publisher

DeepSeek Official spec

Architecture

Sparse mixture of experts Official spec

Total parameters

304.18B Official spec

Active parameters

13B per token Official spec

Context length

1,048,576 tokens Official spec

Attention ⓘ

MQA · 1 KV heads × 512 dim × 43 layers Official spec

Licence

MIT Official spec source

Specification confidence ⓘ

High · verified 2026-09-07 Official spec source

Released

1 May 2026 Official spec

Official source

Hugging Face

Quantizations and memory

Quantization	Format	Bits/weight	Weights	Quality kept
`MLX 4-bit`	mlx	4.5	165.7 GB	98.0%
`Q4_K_M`default	gguf	4.85	176.9 GB	98.5%
`Q8_0`	gguf	8.5	307.0 GB	99.9%

Weight sizes are computed from the parameter count and bits per weight plus a format-specific overhead for the layers that stay at higher precision — not read from a specific published file. Quality retention is an assumption, not a measured evaluation.

Recommendations

Cheapest that can run it

Mac Studio M3 Ultra 256 GB

~51.8–74.5 t/s · €6.299

Lowest purchase price among configurations where the model fits at some quantization in our catalogue. Speed is not considered.

Cheapest above 20 t/s

Mac Studio M3 Ultra 256 GB

~51.8–74.5 t/s · €6.299

20 tokens/second is roughly the point at which generation keeps pace with reading. Below it, interactive use feels like waiting.

Cheapest above 40 t/s

Mac Studio M3 Ultra 256 GB

~51.8–74.5 t/s · €6.299

40 tokens/second is the threshold most people describe as comfortable for coding agents, where output arrives faster than you can review it.

Fastest with real measurements

Nothing in the database qualifies.

Highest throughput among configurations with an actual published measurement rather than our estimate.

Best throughput per euro

Mac Studio M5 Ultra 256 GB

~89.3–129 t/s · €9.599

Highest decode tokens/second per EUR 1,000 of purchase price. Ignores running costs and resale — see the economics section for the full picture.

Hardware that runs DeepSeek-V4 Flash0 measured, 4 estimated

Quantization

Minimum decode: 0 t/s

4 of 4 rows

Hardware↕	Quant	Memory↕	Decode▼	Prefill↕	Context	Price↕	Fit	Confidence↕
Mac Studio M5 Ultra 256 GB Apple · 256 GB · 1,200 GB/s	`MLX 4-bit`	172.4 GB	~89.3–129 t/s	~462–959 t/s	256K	€9.599	Comfortable	Estimated
Mac Studio M3 Ultra 256 GB Apple · 256 GB · 819 GB/s	`MLX 4-bit`	172.4 GB	~51.8–74.5 t/s	~111–231 t/s	256K	€6.299	Comfortable	Estimated
Mac Studio M5 Ultra 512 GB Apple · 512 GB · 1,200 GB/s	`Q8_0`	317.9 GB	~49.8–71.7 t/s	~462–959 t/s	512K	€12.599	Comfortable	Estimated
Mac Studio M3 Ultra 512 GB Apple · 512 GB · 819 GB/s	`Q8_0`	317.9 GB	~28.3–40.7 t/s	~111–231 t/s	512K	€9.199	Comfortable	Estimated

Hosted alternatives for this exact model

These endpoints serve the same open weights, so comparing them against local ownership is a like-for-like economic question rather than a quality trade-off.