---
name: Deepseek-V4-Flash
slug: v4-flash
type: LLM
providers:
  - azure
  - opencode
  - kilo
  - digitalocean
  - nano-gpt
  - umans-ai
  - openrouter
  - umans-ai-coding-plan
  - neuralwatt
  - venice
  - vivgrid
  - ollama-cloud
  - deepseek
  - baseten
  - chutes
  - kenari
description: 'Fast DeepSeek model for efficient chat, coding help, and agent loops'
tags:
  - reasoning
  - open-source
  - tool-use
  - deepseek-flash
  - llm
  - deepseek
is_open_source: false
context_window: 1048576
max_output: 1048576
pricing:
  - image: 0
    input: 0.19
    output: 0.51
    provider: azure
  - image: 0
    input: 0.13
    output: 0.26
    provider: baseten (Ver 0731)
  - image: 0
    input: 0.14
    output: 0.28
    provider: chutes (TEE / Ver 0731)
  - image: 0
    input: 0.14
    output: 0.28
    provider: deepseek
  - image: 0
    input: 0.0679
    output: 0.168
    provider: digitalocean
  - image: 0
    input: 0
    output: 0
    provider: kenari (Free Tier)
  - image: 0
    input: 0.079996
    output: 0.252
    provider: kilo
  - image: 0
    input: 0.2
    output: 0.4
    provider: nano-gpt (TEE)
  - image: 0
    input: 0.14
    output: 0.28
    provider: nano-gpt (Ver 0731)
  - image: 0
    input: 0.14
    output: 0.28
    provider: nano-gpt
  - image: 0
    input: 0.104
    output: 0.207
    provider: neuralwatt
  - image: 0
    input: 0
    output: 0
    provider: opencode (Free Tier)
  - image: 0
    input: 0.14
    output: 0.28
    provider: opencode
  - image: 0
    input: 0.079996
    output: 0.252
    provider: openrouter
  - image: 0
    input: 0
    output: 0
    provider: umans-ai-coding-plan
  - image: 0
    input: 0.14
    output: 0.28
    provider: umans-ai
  - image: 0
    input: 0.35
    output: 0.7
    provider: venice (Ver 0731)
  - image: 0
    input: 0.138
    output: 0.275
    provider: venice (Ver 0423)
  - image: 0
    input: 0.15
    output: 0.3
    provider: vivgrid
  - image: 0
    input: 0.08
    output: 0.18
    provider: openrouter (Ver 0731)
release_date: '2024-01-01'
openrouter_id: ~deepseek/deepseek-v4-flash-latest
capabilities:
  audio: false
  vision: false
  tool_use: true
  citations: false
  reasoning: true
---
Deepseek-V4-Flash is a LLM model.
Fast DeepSeek model for efficient chat, coding help, and agent loops