DeepSeek V4.1 Flash (1017)

deepseek-ai/DeepSeek-V4.1-Flash-1017serverlessdedicated

DeepSeek's fast tier with vision input: 1M context and the best price-to-throughput ratio in the sference catalog for vision-capable workloads.

visionthinkingtool calling
Input
$0.50 / 1M
Output
$1.50 / 1M
Cached
$0.05 / 1M
Context
1M
Type
Chat
Maker
DeepSeek
Released
2026-10-17

Compared to the portfolio

Intelligence vs price
Chat models on sference
Index
45/100
Blended
$0.75/1M
↖ Smarter for less
30
35
40
50
$0.25$4.00
45$0.75
DeepSeek V4.1 Flash (1017)
↑ IndexBlended $ / 1M · log →

Artificial Analysis Intelligence Index v4.3 (0–100), measured across all providers by Artificial Analysis, not sference specifically. Blended price weights input and output 3:1. Hover a point for the reasoning-effort variant scored. Source

ModelContextInputOutputIndexVision
Kimi K31M$3.00$15.0044✓
GLM 5.31M$1.20$4.2045—
DeepSeek V4.1 Flash (1017)1M$0.50$1.5045✓
GLM 5.3 Flash1M$0.20$0.6042✓
DeepSeek V4 Flash (0731)1M$0.28$0.5634—
Clef 27Bdecisions64K$0.24——✓

Serverless rates per 1M tokens at the realtime (list) rate — priority bills 1.3×, flex 0.7×, and batch 0.5× of it — confirmed per model when your account is provisioned. Dedicated deployments are billed per GPU-hour on monthly commitments.

Use it

POST /v1/chat/completions
Realtime, priority, and flex (service_tier)
POST /v1/responses
Realtime, priority, flex, and the 24h batch window (background: true)
POST /v1/messages
Anthropic Messages shape, realtime only
curl https://api.sference.com/v1/chat/completions \
  -H "Authorization: Bearer $SFERENCE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
    "messages": [{ "role": "user", "content": "Say hello in one sentence." }]
  }'

OpenAI-compatible; most clients switch with a base-URL change. Quickstart · Python SDK · Claude Code

Dedicated deployment

Run DeepSeek V4.1 Flash (1017) on GPU capacity reserved for you: a private endpoint, predictable latency, billed per GPU-hour on a monthly commitment. Tell us your expected volume and we come back with a sizing and a quote. How dedicated works

No commitment. Prefer email? [email protected]

FAQ

How is DeepSeek V4.1 Flash different from V4 Flash?

V4.1 Flash is the successor to V4 Flash, first published by DeepSeek on 9 September 2026 and refreshed with the 1017 snapshot on 17 October 2026. It adds image input, which is new on the DeepSeek line, and keeps the 1M context, tool calling, and thinking. Per token it costs more than V4 Flash on input and output but has a lower cached-input rate. V4 Flash keeps running; we will give notice with a date before it retires.

What is the difference between `DeepSeek-V4.1-Flash` and `DeepSeek-V4.1-Flash-1017`?

deepseek-ai/DeepSeek-V4.1-Flash-1017 pins DeepSeek's 17 October 2026 snapshot. deepseek-ai/DeepSeek-V4.1-Flash is an alias that currently serves it. Use the dated id if you want the version you validated to stay fixed; the alias can move to a newer DeepSeek release, which we announce by email first.

Should I move from V4 Flash to V4.1 Flash?

Move if you need image input, or if your prompts reuse long cached prefixes. For text-only, cost-sensitive workloads, V4 Flash remains the cheapest DeepSeek option in the catalog. Rerun your evals on V4.1 before switching; both stay addressable by their pinned ids.

What precision is DeepSeek V4.1 Flash served at?

DeepSeek's native checkpoint, as published: FP8 weights with FP4 routed experts. We add no extra quantization.

What can DeepSeek V4.1 Flash (1017) do on sference?

DeepSeek V4.1 Flash (1017) has a 1M-token context window and supports vision, thinking, and tool calling. Send images as image content parts on the chat, Responses, or Messages endpoints. Turn on thinking with thinking: {"type": "enabled"}.

Which APIs can I call DeepSeek V4.1 Flash (1017) through?

POST /v1/chat/completions and POST /v1/responses (realtime, flex via service_tier: "flex", and the 24h async window), plus POST /v1/messages (Anthropic-shaped, realtime only). The OpenAI and Anthropic SDKs work by changing the base URL, and the same model id works on every endpoint. See the quickstart and the Anthropic Messages guide.

Is DeepSeek V4.1 Flash (1017) cheaper for background or batch work?

Yes. Flex (service_tier: "flex", enabled per account) bills 0.7× the realtime rate for requests that can wait up to 30 minutes, and the 24h batch window (batches and background responses) bills 0.5×. Latency-critical requests can go the other way: service_tier: "priority" jumps the queue at 1.3×. Per 1M tokens: priority $0.65 in / $1.95 out, flex $0.35 in / $1.05 out, batch $0.25 in / $0.75 out. See pricing.

Where is data sent to DeepSeek V4.1 Flash (1017) processed, and is it used for training?

Every request runs on European GPUs. Inputs are never used for training, contractually and technically, and are deleted per your configured retention policy; compliance logs (model version, region, timestamps) are kept for your audit trail. Details in the DPA and sub-processor list.

What are the rate limits for DeepSeek V4.1 Flash (1017)?

There is no general per-account request limit, and the API does not send rate-limit headers. Image input has a per-team budget of 20 images per second; exceeding it returns 429. See Images and vision.

What happens when DeepSeek V4.1 Flash (1017) is updated or retired?

Pin an exact version by calling its versioned id: that version is the one that serves, and every response records it. Before a model retires we email you with a date, and the old id becomes an alias of its successor so existing integrations keep working. Rerun your evals against the successor before the date.

Generated from the live catalog (GET /v1/models), so pricing, context, and capabilities always match the API. Machine-readable version: https://sference.com/models/deepseek-ai/DeepSeek-V4.1-Flash-1017.md