When was DeepSeek V4.1 Flash released?

DeepSeek released DeepSeek V4.1 Flash on 10 Sep 2026. It is available through the Opper gateway.

Does DeepSeek V4.1 Flash support zero data retention?

Yes. DeepSeek V4.1 Flash is available with zero data retention on DeepInfra, Fireworks, GreenPT, Melious, Nextbit, Novita, Sference, Wafer through the Opper gateway. Other routes carry zero data retention with abuse monitoring; see each route's window in the data-handling section.

Where is DeepSeek V4.1 Flash hosted?

DeepSeek V4.1 Flash runs on both EU and US hosting routes through the Opper gateway. Pick the route that matches your residency requirement.

Is DeepSeek V4.1 Flash trained on prompts and outputs?

No. Every hosting route for DeepSeek V4.1 Flash is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for DeepSeek V4.1 Flash?

Yes. A GDPR Data Processing Agreement is available on at least one route serving DeepSeek V4.1 Flash. Contact hello@opper.ai to receive it.

DeepSeek V4.1 Flash

by DeepSeek

Released 28 days after DeepSeek V4 Pro · All DeepSeek releases

DeepSeek-V4.1-Flash, released on September 10, 2026, is the smallest model in DeepSeek's new architecture family and brings native visual understanding to the Flash tier. It carries 552 billion backbone parameters but activates only 8 billion per token during prefill and 16 billion during decode, supports a context window of up to one million tokens, and was trained from scratch on a 45-trillion-token multimodal corpus. The Causal Encoder-Decoder design arranges its 40 Transformer layers as a 20-layer causal encoder feeding a 20-layer decoder, and Compressed Sparse Attention 2 with FP4 main KV caching brings the global KV-cache footprint down to 890 bytes per token, roughly a fourfold reduction against V4 Flash. Reasoning effort is continuously controllable as an integer from 1 to 100 rather than the three fixed levels of the V4 generation. DeepSeek reports 90.6 on Terminal-Bench 2.1 against 82.7 for the July V4 Flash checkpoint, 74.2 on DeepSWE v1.1, 90.9 on GPQA Diamond, 88.1 on CyberGym, and a Codeforces rating of 3471. Weights are published under the MIT license. It fits agentic coding, long-context document and image work, and high-throughput agent loops that want a strong model with a small per-token compute budget.

Call DeepSeek V4.1 Flash on Opper with the OpenAI SDK. Sign up without a credit card. Get started

Get your API key Try in chat Compare

Key info

Input

Output

Features

Context window

Max output

393K

Input price

$0.30 /1M$0.0060 cache read

Output price

$1.20 /1M

Released

10 Sep 2026

EU residency available
US residency available
Zero data retention on pay-as-you-go
No training by default
GDPR DPA available

Available routes

DeepSeek V4.1 Flash runs on 12 different routes through the Opper gateway. Compare residency, zero data retention and training posture at a glance, with full data-handling detail per route below. In the EU it is served by GreenPT, Melious (Germany) and Sference. See every DeepSeek model hosted in the EU.

Provider	Region	Training	Input	Output	Cache read
BytePlus	Multi	No	$0.30	$1.20	$0.0060
DeepInfra	US	No	$0.20	$0.60	$0.0060
Fireworks	Multi	No	$0.22	$0.66	$0.0070
GreenPT	EU	No	$0.25	$1.25	$0.01
Melious	EU	No	$0.23	$1.14	$0.01
Morph	US	No	$0.30	$1.20	$0.03
Nebius	Multi	No	$0.30	$1.20	Input price
Nextbit	Multi	No	$0.21	$0.84	$0.0040
Novita	US	No	$0.30	$1.20	$0.0060
Sference	EU	No	$0.50	$1.50	$0.05
Tencent Cloud	Multi	No	$0.30	$1.20	$0.0060
Wafer	US	No	$0.30	$1.20	$0.01

Uptime and availability

DeepSeek V4.1 Flash runs on 4 independently monitored routes through the Opper gateway. A direct integration leaves you on one of them, the gateway serves whichever is healthy.

100%effective uptime, last 30 days

Averaged across these routes on their own, a single provider reached 99.55% over the same window.

Daily status through the gateway, which serves each request from whichever route is healthy.

4 monitored providers serve DeepSeek V4.1 Flash.

Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. 8 further routes for this model are not on the board yet and are not counted above. See uptime for every provider Opper monitors.

Data handling per route

Each route hosting DeepSeek V4.1 Flash has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

BytePlus — Multi-region

Zero data retention with abuse monitoring: prompts and outputs are kept for up to 180 days for abuse monitoring only, and are not used for training. No training on customer data. GLOBAL; SCCs; DPA available.

Zero data retention: With abuse monitoring, 180-day window.
Training: No training on customer data.
Logging: Abuse monitoring (180-day retention)
Abuse monitoring: On by default, holds flagged content, 180 days, stored in Malaysia, a person can view samples
Caching: Content cached for replay
GDPR DPA: DPA available
Transfer mechanism: SCCs

DeepInfra — United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; unknown; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: No classifier
GDPR DPA: DPA available
Transfer mechanism: unknown

Fireworks — Multi-region

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. GLOBAL; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: On by default, keeps nothing
Subprocessor access: Subprocessors may read content
GDPR DPA: DPA available
Transfer mechanism: SCCs

GreenPT — EU🇪🇺

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. EU; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: Keeps nothing
Caching: Content cached for replay
Subprocessor access: Subprocessors may read content
GDPR DPA: DPA available
Transfer mechanism: Not applicable — data stays in EU

Melious — Germany🇩🇪

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. EU; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: No classifier
Subprocessor access: Subprocessors may read content
GDPR DPA: DPA available
Transfer mechanism: Not applicable — data stays in EU

Morph — United States🇺🇸

Zero data retention with abuse monitoring: prompts and outputs are kept for up to 30 days for abuse monitoring only, and are not used for training. No training on customer data. US; SCCs; DPA available.

Zero data retention: With abuse monitoring, 30-day window.
Training: No training on customer data.
Logging: Abuse monitoring (30-day retention)
Abuse monitoring: On by default, holds flagged content, a person can view samples
GDPR DPA: DPA available
Transfer mechanism: SCCs

Nebius — Multi-region

Content is retained: provider logging. No training on customer data. GLOBAL; SCCs; DPA available.

Zero data retention: Retained: provider logging.
Training: No training on customer data.
Logging: Other
Abuse monitoring: No classifier
GDPR DPA: DPA available
Transfer mechanism: SCCs

Nextbit — Multi-region

Zero data retention: nothing is logged or used for training. No training on customer data. GLOBAL; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged or used for training.
Training: No training on customer data.
Logging: None
Caching: Content cached for replay
Subprocessor access: Subprocessors may read content
GDPR DPA: DPA available
Transfer mechanism: SCCs

Novita — United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: On by default, keeps nothing
GDPR DPA: DPA available
Transfer mechanism: SCCs

Sference — EU🇪🇺

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. EEA; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: No classifier
Caching: Content cached for replay
Subprocessor access: No subprocessor reads content
GDPR DPA: DPA available
Transfer mechanism: SCCs

Tencent Cloud — Multi-region

Content is retained: limited debug logs, 30-day retention. No training on customer data. GLOBAL; SCCs; DPA available.

Zero data retention: Retained: limited debug logs, 30-day retention.
Training: No training on customer data.
Logging: Limited debug logs (30-day retention)
Abuse monitoring: On by default, holds flagged content
Caching: Content cached for replay
GDPR DPA: DPA available
Transfer mechanism: SCCs

Wafer — United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: No classifier
GDPR DPA: DPA available
Transfer mechanism: SCCs

Benchmarks

Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.

Global rank#60 of 676 LLMs

TierStrong

Output speed220 tok/s

First token0.82s

Intelligence Index39.5

Reasoning & knowledge

Humanity's Last Exam

39%

Long-context reasoning

84%

Coding

SciCode

52%

See full leaderboard →Benchmarks via Artificial Analysis · View on AA

Get started

Call DeepSeek V4.1 Flash through the Opper gateway with one API key. Opper is drop-in compatible with the OpenAI, Anthropic and Google AI SDKs.

DeepSeek V4.1 Flash is a premium model on Opper. Sign up needs no credit card: you get an API key straight away and the free models work in the playground and the API. Add a card to use premium models, pay-as-you-go with no minimum.

Set it up with your agent

Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up Opper for you.

Or call it directly

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.OPPER_API_KEY,
5  baseURL: "https://api.opper.ai/v3/compat",
6});
7
8const completion = await client.chat.completions.create({
9  model: "deepseek-v4.1-flash",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);