CrossModel
DeepSeek V4 Flash
deepseek/deepseek-v4-flash
Back to model catalog

DeepSeek · Model guide

DeepSeek V4 Flash

Compare
deepseek/deepseek-v4-flash10% off

A 284B-parameter sparse MoE model with 13B active parameters and a 1M-token context window. It is the economical DeepSeek V4 option for coding, search and tool-using agents, with an optional Max reasoning mode for harder work.

Modalities
TextText
Context
1M
Max output
384K
Price / 1M tokens
$ 0.45$ 0.41in$ 1.22out
StreamingTool callingJSON outputImage inputThinking mode · can be turned off

What this model is good at

What the vendor positions it for, and which of our models to reach for instead.

VendorThe more economical V4 model, intended for cost-sensitive and simpler agent workloads.source

Best for

  • Cost-sensitive coding agents · Strong software-engineering scores with lower serving cost than Pro.(vendor claim)source

  • High-volume tool workflows · Flash is positioned for simpler agents and cost-sensitive production workloads.(vendor claim)source

Neighbouring models
Compare them side by side →

Pricing and billing

Billed per token. Cached input is charged at the cache-read rate.

Price / 1M tokensYou pay
Input$ 0.45$ 0.41
Output$ 1.35$ 1.22
Cache read$ 0.015$ 0.013
Cache write$ 0.45$ 0.41
Estimate a request
$ 0.0065
Estimated cost per request
10.0K × $ 0.41
+ 2.0K × $ 1.22
≈ $ 6.48 per 1,000 requests

Capabilities and limits

What CrossModel guarantees across every route this model can take right now.

Context window
1.0M tokens
Max output
384.0K tokens
Input / output modalities
Text → Text
Streaming
Supported
Tool calling
Supported
Structured output (JSON)
Supported
Image input
Not supported
Thinking mode
Supported· can be turned off
Reasoning effort
low · medium · high · xhigh · max
Available endpoints
/v1/chat/completions · /v1/responses · /v1/messages
First-party sources:Specshuggingface.co 1api-docs.deepseek.com 2api-docs.deepseek.com 3Capabilitiesapi-docs.deepseek.com 1api-docs.deepseek.com 2api-docs.deepseek.com 3api-docs.deepseek.com 4api-docs.deepseek.com 5

Published benchmarks

Scores the vendor reported, with the evaluation setup each one came from.

Agentic coding
  • Terminal-Bench 2.1
    pass_rate·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立运行；不可与 DeepSeek Harness 的厂商成绩直接互换。

    78.7%
Coding
  • SciCode
    score·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立评测。

    49.9%
Knowledge work
  • GDPval-AA v2
    elo·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立评测；记录页面展示的原始 Elo。

    1558.4
Long context
  • AA-LCR
    score·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立长上下文推理评测。

    74.3%
Reasoning
  • AA Intelligence Index v4.1.1
    score·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立评测的九项复合指数，不是百分比；来源页未披露各项完整可复现设置。

    51.8
  • Humanity's Last Exam
    accuracy·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立评测。

    38.6%
Scientific reasoning
  • CritPt
    score·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立物理推理评测。

    16.6%
  • GPQA Diamond
    accuracy·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立评测。

    90.8%
Tool use
  • tau3-Banking
    score·Artificial Analysis · 2026-08-13effort maxsetup not fully disclosed

    Artificial Analysis 独立评测。

    39.4%

Vendor-reported numbers, not CrossModel measurements. Scores are only comparable when the benchmark version, metric and evaluation setup match, so nothing here is averaged or ranked.

Use it in your tools

Point the base URL at CrossModel and paste this model ID — every tool below has a setup guide.

Claude Code
Anthropic's official command-line coding assistant
Cursor
AI-first code editor
Cline
Open-source AI coding agent for VS Code
Codex
OpenAI's coding agent for ChatGPT desktop and the terminal
OpenCode
Open-source terminal coding assistant
Zed
High-performance, AI-native code editor
Open WebUI
Open-source, self-hosted AI chat interface
Continue
Open-source AI coding assistant for VS Code & JetBrains
See all integration guides →

Frequently asked questions

What is DeepSeek V4 Flash?
DeepSeek V4 Flash is available on CrossModel as deepseek/deepseek-v4-flash. It has a 1M-token context window and can return up to 384K tokens per request.
How much does DeepSeek V4 Flash cost?
Currently $ 0.41 per 1M input tokens and $ 1.22 per 1M output tokens. Cached input is billed at $ 0.013 per 1M.
Does DeepSeek V4 Flash support tool calling and structured output?
Tool calling: Supported · Structured output (JSON): Supported · Image input: Not supported · Streaming: Supported · Thinking mode: Supported
Which endpoint do I call?
OpenAI Chat Completions (/v1/chat/completions), OpenAI Responses (/v1/responses) and Anthropic Messages (/v1/messages) all work — pick whichever your SDK already speaks.
Can I try it without writing code?
Yes. Sign in and open the Playground on this page to chat with the model directly — no API key needed. Usage is billed from your wallet at the normal rate.

Chat first, integrate later

Chat first, then wire it in once you like the answers.