DeepSeek

DeepSeek V4 FlashOpen

Name: DeepSeek V4 Flash
Price: 0.28 USD
Rating: 42 (11 reviews)
Author: DeepSeek

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.

#403released Jul 30, 2026·Open Weights·Not disclosed

text1.0M ctx

API providers Get API Key →Leaderboard

Intelligence

42.0 #403

AAQI 50.0

Output Speed

113 t/s

TTFT 1.34 s · ITL 8.8 ms/tok

Blended price (3:1)

$0.175 /1M

In $0.14/M · Out $0.28/M

Context

1.0M

≈ 2,500 A4 pages · max out 16,384

Quality Domain Radar

6-Axis Frontier Evaluation vs Category Average & Flagship

DeepSeek V4 Flash

Claude Mythos Preview (#1)

Average (68)

Radar axes are approximated from category scores where direct benchmark measurements are missing — see Methodology.

API deployment

Compare hosts & latency

Throughput, TTFT and cache pricing across Groq, Cerebras, Together, DeepInfra, Fireworks, Bedrock and Azure.

GroqCerebrasTogether AIDeepInfraFireworks AIAWS BedrockAzure AI

View providers matrix →

Scores by Source

Arena.ai ↗

1438.0

#79 · 48.4k votes · ±4

Artificial Analysis ↗

Category Scores

Value100.0 Long Context99.2 Math62.0 Intelligence51.8 Reasoning41.2 Coding38.0 Agentic35.3 Speed22.0 Knowledge12.5

Benchmark Results

LiveCodeBench91.6%GPQA Diamond88.1%MMLU-Pro86.2%SWE-Bench Verified79%

Browsecomp73.2%

MCP Atlas69%SWE-Bench Pro52.6%SciCode49.9%Toolathlon47.8%Humanity’s Last Exam45.1%

Model Details

Release date

Jul 30, 2026

Parameters

Not disclosed

License

Open Weights

Context Window

1,000,000 tokens

Input Price

$0.14/M

Output Price

$0.28/M

Output Speed

113 t/s

Time to First Token

1.34 s

Estimated monthly bill

Monthly API requests: 100,000Avg input tokens: 1,500Avg output tokens: 500Prompt cache hit rate: 50%

Estimated monthly spend

$0.00

Calculating savings vs GPT-4o...

Cite & embed

GitHub Markdown badge

Loading badge markdown...

BibTeX

@misc{llmpodium2026deepseekv4flash,
  title = {Benchmark Analysis and Intelligence Evaluation of DeepSeek V4 Flash},
  author = {LLMPodium Research},
  year = {2026},
  url = {https://llmpodium.com/models/deepseek-v4-flash}
}