Requesty
DeepInfra Inc. logo
DeepInfra Inc.

deepseek-ai/DeepSeek-V4-Flash

DeepSeek V4 Flash is an efficiency-focused MoE model with 284B total parameters (13B active) and a 1M-token context window. It's tuned for fast inference and high-throughput use cases while still holding up on reasoning and coding tasks.

ReasoningTool callingCachingJSON schema
CompareDocs

Specifications

Context window1.0M tokens
Max outputN/A
API typechat
AddedMay 27, 2026
Model IDdeepinfra/deepseek-ai/DeepSeek-V4-Flash
Data retentionNo
Used for trainingNo
Provider location🇺🇸 US
Privacy policyDeepInfra Privacy Policy→

Benchmarks

Released 2026-04-24
Coding Indexcoding
56.2%

Artificial Analysis Coding Index — a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.

GPQA Diamondreasoning
89.4%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
40.3%

Artificial Analysis Intelligence Index — a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality. Always test on your own workload.

Pricing

Input / 1M
$0.10
Output / 1M
$0.20
Cache write
N/A
Cache read / 1M
$0.02
Estimated cost
100K input + 10K output$0.0120
1M input + 100K output$0.12
10M input + 1M output$1.20

Requesty charges exactly what the upstream provider charges, no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to deepinfra/deepseek-ai/DeepSeek-V4-Flash.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="deepinfra/deepseek-ai/DeepSeek-V4-Flash",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other DeepInfra Inc. models

zai-org/GLM-5.1
203K ctx$1.05/1M
Coding 56
nvidia/Nemotron-3-Nano-30B-A3B
262K ctx$0.05/1M
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B
262K ctx$0.10/1M
Coding 38
moonshotai/Kimi-K2.6
262K ctx$0.75/1M
Coding 62
gemma-4-31B-it
262K ctx$0.13/1M
deepseek-ai/DeepSeek-V4-Pro
1.0M ctx$1.30/1M
Coding 59

Frequently asked questions

How much does deepseek-ai/DeepSeek-V4-Flash cost?
deepseek-ai/DeepSeek-V4-Flash is priced at $0.10 per million input tokens and $0.20 per million output tokens when accessed via Requesty. Prompt caching is supported, which can cut effective input cost by up to 90% on repeated context. Requesty charges exactly what the upstream provider charges, we don't add markup.
What is the context window of deepseek-ai/DeepSeek-V4-Flash?
deepseek-ai/DeepSeek-V4-Flash has a context window of 1.0M tokens. That's roughly 1,398 words of input you can fit in a single prompt.
How does deepseek-ai/DeepSeek-V4-Flash perform on benchmarks?
deepseek-ai/DeepSeek-V4-Flash scores 95.0% on τ²-Bench, 89.4% on GPQA Diamond, 56.2% on Coding Index. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can deepseek-ai/DeepSeek-V4-Flash do?
deepseek-ai/DeepSeek-V4-Flash supports tool calling, extended reasoning, prompt caching, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use deepseek-ai/DeepSeek-V4-Flash with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "deepinfra/deepseek-ai/DeepSeek-V4-Flash". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run deepseek-ai/DeepSeek-V4-Flash through Requesty?
Yes. deepseek-ai/DeepSeek-V4-Flash runs through Requesty's OpenAI-compatible API, served from DeepInfra Inc.. You do not host the model yourself: point base_url at Requesty, set the model to "deepinfra/deepseek-ai/DeepSeek-V4-Flash", and requests are routed to the upstream provider with automatic failover. The same key gives you 400+ other models too.

Access deepseek-ai/DeepSeek-V4-Flash through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

All DeepInfra Inc. models