DeepSeek

DeepSeek V4 Flash 0731

Launched 2026-07-31 · Knowledge up to 2025-05 · Listed on OpenRouter 2026-07-31

Prices updated . Benchmarks updated .

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....

You send / 1M

$0.060

tokens in

It replies / 1M

$0.180

tokens out

Context

1M

tokens it can read at once

Typical request

$0.0002

1,000 words in, 400 out

Benchmarks

BenchmarkScoreKindDateSource
Epoch Capabilities Index154.9Independent2026-07-31Epoch AI
GPQA DiamondMax effort91.0%Independent2026-08-02Epoch AI
OTIS Mock AIMEMax effort94.4%Independent2026-08-02Epoch AI
SimpleQA VerifiedMax effort33.6%Independent2026-08-27Epoch AI
ARC-AGI-2Max effort61.4%Independent2026-07-31ARC Prize, republished by Epoch AI
CritPtMax effort16.6%Independent2026-07-31CritPt, 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
Relace · fp4$0.070$0.140100%Read
DeepInfra · fp8$0.048$0.14497%Read
DeepInfra · fp8$0.060$0.180100%Read
Wafer · nvfp4$0.140$0.230100%Read
DigitalOcean$0.119$0.238100%Read
BaseTen · fp8$0.130$0.260100%Read
BaseTen · fp32$0.130$0.260100%Read
Venice$0.130$0.260100%Read
DeepInfra · fp8$0.090$0.270100%Read
CoreWeave · fp8$0.130$0.270100%Read
Cohere · fp8$0.140$0.280100%Read
Parasail · fp8$0.140$0.28099%Read
Wafer · nvfp4$0.043$0.300100%Read
Sail Research · fp4$0.150$0.30099%Read
Sail Research · fp4$0.150$0.30099%Read
Reka$0.100$0.420100%Read
Inceptron · fp4$0.130$0.495100%Read
Mancer 2 · fp8$0.150$0.50098%Read
SiliconFlow · fp8$0.220$0.66099%Read
Novita · fp8$0.409$1.23100%Read
PrimeIntellect · fp4$0.440$1.32100%Read
Cloudflare$0.440$1.32100%Read

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

What it can do

  • Reads: text
  • Writes: text
  • Weights: Open weights
  • Reasoning: Yes
  • Tool calling: Yes
  • Longest reply: 393K tokens
  • Repeated text is billed at $0.015 per 1M (prompt caching)

Compare with

Appears in

Estimate your monthly cost with DeepSeek V4 Flash 0731

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.