DeepSeek-V4.1-Flash

DeepSeek · released 2026-09-10 · MIT (open-weight, per Hugging Face model card) · announcement · weights

Price

Input $0.3/1M

Cached read $0.006/1M

Output $1.2/1M

$ / 1M tokens, primary tier. full pricing reference

priced on 2 hosts

Context window

1,000,000 tokens

Rate limits

not published as a table by DeepSeek.

Retirement

no retirement announced.

Runs on

vLLM · PR #67378 · merged 2026-11-13
vLLM · PR #64715 · merged 2026-11-10
vLLM · PR #64589 · merged 2026-11-02
vLLM · PR #64804 · merged 2026-10-31
SGLang · PR #46668 · merged 2026-10-26
vLLM · PR #62384 · merged 2026-10-23
vLLM · PR #62183 · merged 2026-10-19
SGLang · PR #44937 · merged 2026-10-18
vLLM · PR #59857 · merged 2026-10-16
SGLang · PR #44654 · merged 2026-10-16
SGLang · PR #41476 · merged 2026-10-04
vLLM · PR #57407 · merged 2026-09-27
SGLang · PR #40532 · merged 2026-09-21
vLLM · PR #56227 · merged 2026-09-18
Ollama · ollama pull deepseek-v4.1-flash · library entry

Latest mentions

TOOLING [ROCm][Docs] Add DeepSeek-V4.1-Flash MI355X DP2 AgentX preset with UMBP KV offload — · 25h ago
TOOLING [ROCm][UMBP] Support UMBP standalone mode for DeepSeek-V4.1-Flash — · 4d ago
TOOLING Lawyer with too many GPUs, part 4: DeepSeek-V4.1-Flash on eight V100s from 2018. 200+ tok/s on code, ~110 on legal writing, 2,200 tok/s prefill at 128K, 524K context, every weight in VRAM — fuck you sam altman. · 8d ago
TOOLING DeepSeek-V4.1-Flash-1017 on one RTX 3090: Big updates to the Delta Inference Engine v0.93. Same 41 tok/s at 192GB/IQ2 at better quality, usable from 64 GB RAM, first token ~40% sooner, new Unsloth quants at full speed, EXL3, basic multi-GPU, public CI and more — Big update to the Delta Inference Engine, you guys asked for more features, more benchmarks and less clickbait. · 8d ago
TOOLING [Docs] Add DeepSeek-V4.1-Flash MI325X AgentX preset — · 11d ago
TOOLING [Docs][ROCm] Add DeepSeek-V4.1-Flash MI355X TP8 concurrency-1 AgentX preset — · 13d ago
RELEASE Introducing DeepSeek-V4.1-Flash-1017 Dynamic v3.1 Unsloth GGUFs — Hey everyone! We’re releasing new DeepSeek-V4.1-1017 Dynamic V3.1 (preview) GGUFs which showcases how Unsloth’s GGUFs retain the most… · 15d ago
TOOLING [Doc] sgl-router: Add GLM-5.3-Flash and DeepSeek-V4.1-Flash deployment examples — · 18d ago
TOOLING [Feature][PCP] Support decoder-side SWA bounded replay with APC for DeepSeek-V4.1-Flash — · 21d ago
TOOLING [Feature][PCP] Support APC and encoder-side SWA bounded replay for DeepSeek-V4.1-Flash — · 26d ago
RELEASE DeepSeek: DeepSeek v4.1 Flash 1017 — DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model… · 28d ago
TOOLING [Feature][PCP] Add DeepSeek-V4.1-Flash support — · 29d ago
TOOLING [Doc] cookbook: Add SemiAnalysis AgentX launches to Muse Spark Open, refresh DeepSeek-V4.1-Flash — · 29d ago
TOOLING [AMD] Add DeepSeek-V4.1-Flash MI35x nightly accuracy test — · 41d ago
TOOLING [ROCm][Perf] Enable layer-aware CSA2 multi-stream overlap for DeepSeek-V4.1-Flash — · 47d ago
TOOLING [DSV4.1] Add encoder cuda graph support for deepseek-v4.1-flash — · 55d ago
TOOLING [Feat][Model] Support encoder-side SWA-bounded replay for DeepSeek-V4.1-Flash — · 56d ago