DeepSeek

DeepSeek V4.1 Flash 1017

Launched 2026-10-17 · Knowledge up to 2025-05 · Listed on OpenRouter 2026-10-17

Prices updated . Benchmarks updated .

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 name. V4.1 Flash is a sparse mixture-of-experts model...

You send / 1M

$0.036

tokens in

It replies / 1M

$0.500

tokens out

Context

1M

tokens it can read at once

Typical request

$0.0003

1,000 words in, 400 out

Benchmarks

BenchmarkScoreKindDateSource
Epoch Capabilities Index158.8Independent2026-10-17Epoch AI
GPQA DiamondMax effort93.2%Independent2026-10-19Epoch AI
OTIS Mock AIMEMax effort99.4%Independent2026-10-19Epoch AI
SimpleQA VerifiedMax effort42.5%Independent2026-10-19Epoch AI
ARC-AGI-2Max effort78.1%Independent2026-10-17ARC Prize, republished by Epoch AI
CritPtMax effort24.0%Independent2026-10-17CritPt, republished by Epoch AI

Rankings on this site use the Epoch Capabilities Index, an independent composite. Lab-reported rows are the lab's own figures and are kept separate.

Data privacy

No training or retention: Yes. At least one provider serves it with zero data retention.

ProviderYou send / 1MIt replies / 1MUptimePrivacy policy
Sail Research · fp4$0.040$0.07091%Read
Sail Research · fp4$0.040$0.07071%Read
Ionstream$0.077$0.233100%Read
DeepInfra · fp8$0.078$0.23544%Read
DeepInfra · fp8$0.098$0.294100%Read
Particle AI$0.040$0.300100%Read
River$0.086$0.346100%Read
Makora · fp8$0.120$0.400100%n/a
StepFun · fp8$0.105$0.420100%n/a
DeepInfra · fp8$0.147$0.44199%Read
Novita · fp8$0.120$0.480100%Read
Cohere · fp8$0.150$0.500100%Read
Wafer · mxfp4$0.036$0.500100%Read
Wafer · mxfp4$0.070$0.500100%Read
Relace · fp4$0.120$0.500100%Read
Morph · fp8$0.059$0.520100%Read
SiliconFlow · fp8$0.300$0.550100%Read
Particle AI$0.080$0.600100%Read
Cloudflare$0.100$0.600100%Read
InferenceNet · fp8$0.080$0.605100%Read
InferenceNet · fp8$0.080$0.605100%Read
Ionstream$0.210$0.620100%Read
ModelRun · fp4$0.170$0.670100%Read
DekaLLM$0.180$0.68098%Read
PrimeIntellect$0.044$0.70098%Read
DigitalOcean$0.180$0.720100%Read
Fireworks$0.210$0.84088%Read
PrimaLabs$0.240$0.96098%Read
Modal$0.250$1.00100%Read
Nebius · fp8$0.300$1.20100%Read
Mistral · fp8$0.300$1.20100%Read
CoreWeave · fp8$0.300$1.20100%Read
Near AI$0.300$1.20100%Read
Venice · fp8$0.300$1.20100%Read
OpenInference · fp4$0.090$1.20100%Read
Together$0.300$1.20100%Read
Crusoe · fp8$0.290$1.20100%Read
Fireworks$0.300$1.20100%Read
BaseTen$0.300$1.20100%Read
Parasail · fp8$0.300$1.20100%Read
Fireworks$0.450$1.80100%Read
Morph · fp8$0.600$2.4099%Read
BaseTen$0.600$2.40100%Read

Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.

What it can do

  • Reads: text, image
  • Writes: text
  • Weights: Open weights
  • Reasoning: Yes
  • Tool calling: Yes
  • Longest reply: 262K tokens
  • Repeated text is billed at $0.010 per 1M (prompt caching)

Compare with

Appears in

Estimate your monthly cost with DeepSeek V4.1 Flash 1017

Sources and licenses

  • OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
  • Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
  • models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
  • Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
  • LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
  • Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
  • What a score measures: the benchmark guide.