impulse

Fw Deepseek V4 Flash

Open weights
API id FW-DeepSeek-V4-Flash
Input / 1M
$0.15
at Azure AI Foundry
Output / 1M
$0.31
at Azure AI Foundry
Context window
1M
tokens
Max output
384K
tokens per reply
Hosts
1
1 listings
Released
-
date not published

Pricing

List prices at Azure AI Foundry, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.15$0.31
Cached input
80% off input
$0.03

What it costs a month

At Azure AI Foundry list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$57.45
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$54.36
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$97.80
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$434
Use your own numbers

Where to run it

1 host serve Fw Deepseek V4 Flash. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
FW-DeepSeek-V4-Flash
$0.15 / $0.31

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Features
Prompt caching
Open weights
License: DeepSeek License (MIT code)
Runtimes: SGLang, TGI, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than Azure AI Foundry.

Smaller Independent models

No cheaper Independent model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • Qwen3.8 Flash
    Alibaba Qwen · on AIHubMix
    $0.113 / $0.38
    Has every key capability
    6% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    6% less
  • DeepSeek V4.1 Flash
    DeepSeek · on Sail
    $0.09 / $0.30
    Has every key capability
    25% less
  • MiMo V2.6 Flash
    Xiaomi · on GMI Cloud
    $0.112 / $0.224
    Has every key capability
    26% less

Model line

Newer and older versions, at each lab's own price.

  1. UndatedFw Deepseek V4.1 FlashLatest$0.375 / $1.50
  2. UndatedFw Deepseek V4 FlashThis page$0.15 / $0.31

Call it

Using Azure AI Foundry's API.

Python
from litellm import completion

response = completion(
    model="azure_ai/FW-DeepSeek-V4-Flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Fw Deepseek V4 Flash cost?

On Azure AI Foundry, Fw Deepseek V4 Flash costs $0.15 per million input tokens and $0.31 per million output tokens. Cached input costs $0.03 per million tokens.

Which APIs offer Fw Deepseek V4 Flash?

Fw Deepseek V4 Flash runs on 1 API: Azure AI Foundry.

What is Fw Deepseek V4 Flash's context window?

1M tokens, with up to 384K tokens in one reply.

Is Fw Deepseek V4 Flash being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Fw Deepseek V4 Flash?

Yes. Fw Deepseek V4.1 Flash is the newest model in this line.

What are cheaper alternatives to Fw Deepseek V4 Flash?

Models that cost less and keep most of what it can do: Qwen3.8 Flash on AIHubMix (6% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Fw Deepseek V4 Flash's API model id?

Use FW-DeepSeek-V4-Flash with Azure AI Foundry.

Sources

Raw catalog entry for FW-DeepSeek-V4-Flash on Azure AI Foundry
{
  "cache_read_input_token_cost": 3e-8,
  "input_cost_per_token": 1.5e-7,
  "litellm_provider": "azure_ai",
  "max_input_tokens": 1000000,
  "max_output_tokens": 384000,
  "mode": "chat",
  "output_cost_per_token": 3.1e-7,
  "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'",
  "supports_function_calling": true,
  "supports_pdf_input": false,
  "supports_prompt_caching": true,
  "supports_reasoning": true,
  "supports_tool_choice": true
}