Where is the DeepSeek V4 Flash 0423 documentation?

The DeepSeek V4 Flash 0423 documentation on OpenRouter includes the Does DeepSeek V4 Have Vision? blog post, the Why Use OpenRouter for DeepSeek blog post, and the DeepSeek V4 Is Earning Agentic Token Share blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.

How much does DeepSeek V4 Flash 0423 cost?

DeepSeek V4 Flash 0423 costs $0.014/M input tokens and $0.45/M output tokens, with separate rates for Cache Read at $0.014/M tokens.

What is the context length of DeepSeek V4 Flash 0423?

DeepSeek V4 Flash 0423 has a 1,048,576 token context window. It supports up to 1,048,576 completion tokens.

Does DeepSeek V4 Flash 0423 support tool calling and structured outputs?

Yes. DeepSeek V4 Flash 0423 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Which providers serve DeepSeek V4 Flash 0423?

DeepSeek V4 Flash 0423 is served by 13 providers on OpenRouter: Relace, Baidu Qianfan, StreamLake, DeepInfra, GMICloud, Venice, Fireworks, Wafer and 5 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was DeepSeek V4 Flash 0423 released?

DeepSeek V4 Flash 0423 was released on April 24, 2026.

DeepSeek: DeepSeek V4 Flash 0423

deepseek/deepseek-v4-flash

Model weights

Compare

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and high-throughput workloads, while maintaining strong reasoning and coding performance.

The model includes hybrid attention for efficient long-context processing. Reasoning efforts high and xhigh are supported; xhigh maps to max reasoning. It is well suited for applications such as coding assistants, chat systems, and agent workflows where responsiveness and cost efficiency are important.

Modalities

In / out price

$0.014 / $0.45per 1M

Context

1.0M

Released

Apr 24, 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


90% off	$0.14$0.0147	$0.28$0.0294	$0.028$0.00294	0.83s	83 tok/s	86.76%
	$0.12	$0.17	$0.024	0.31s	103 tok/s	100.00%
30% off	$0.138$0.0966	$0.275$0.1925	$0.028$0.0196	0.68s	112 tok/s	100.00%
	$0.014	$0.45	$0.014	0.46s	135 tok/s	99.95%
35% off	$0.14$0.091	$0.28$0.182	$0.028$0.0182	0.81s	137 tok/s	99.77%
	$0.14	$0.28	$0.07	0.43s	117 tok/s	97.15%
	$0.13	$0.28	$0.028	0.95s	62 tok/s	99.01%
	$0.21	$0.56	$0.031	1.14s	114 tok/s	97.33%
	$0.15	$0.50	$0.075	0.72s	103 tok/s	99.78%
	$0.09	$0.18	$0.018	0.37s	121 tok/s	100.00%
	$0.105	$0.21	$0.021	0.58s	103 tok/s	81.79%
80% off	$0.14$0.028	$0.28$0.056	$0.028$0.0056	7.22s	14 tok/s	99.97%
	$0.14	$0.28	$0.028	7.38s	28 tok/s	93.59%

Input $0.0147Output $0.0294
90% off
83 tok/s86.76% uptime
Input $0.12Output $0.17
103 tok/s100.00% uptime
Input $0.0966Output $0.1925
30% off
112 tok/s100.00% uptime
Input $0.014Output $0.45
135 tok/s99.95% uptime
Input $0.091Output $0.182
35% off
137 tok/s99.77% uptime
Input $0.14Output $0.28
117 tok/s97.15% uptime
Input $0.13Output $0.28
62 tok/s99.01% uptime
Input $0.21Output $0.56
114 tok/s97.33% uptime
Input $0.15Output $0.50
103 tok/s99.78% uptime
Input $0.09Output $0.18
121 tok/s100.00% uptime
Input $0.105Output $0.21
103 tok/s81.79% uptime
Input $0.028Output $0.056
80% off
14 tok/s99.97% uptime
Input $0.14Output $0.28
28 tok/s93.59% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.03628/M tokens

Output

$0.2097/M tokens

DeepSeek V4 Flash 0423 — Price history

Price history

No effective price data for this selection

$0.0114	$0.0448	64.7%	18.9%
$0.0188	$0.4493	75.4%	19.1%
$0.0216	$0.0681	47.8%	11.9%
$0.0411	$0.1787	67.8%	11.3%
$0.0434	$0.1827	65.5%	6.2%
$0.0454	$0.1916	66.4%	12.7%
$0.0458	$0.2795	82.5%	1.8%
$0.0580	$0.1684	64.5%	7.9%
$0.0742	$0.2091	36.6%	0.5%
$0.0937	$0.5584	64.9%	0.3%
$0.0958	$0.2783	62.9%	8.7%
$0.1020	$0.4987	63.7%	0.6%
$0.1065	$0.2798	29.8%	0.1%

Effective in $0.0114Effective out $0.0448
Token share
18.9%
Effective in $0.0188Effective out $0.4493
Token share
19.1%
Effective in $0.0216Effective out $0.0681
Token share
11.9%
Effective in $0.0411Effective out $0.1787
Token share
11.3%
Effective in $0.0434Effective out $0.1827
Token share
6.2%
Effective in $0.0454Effective out $0.1916
Token share
12.7%
Effective in $0.0458Effective out $0.2795
Token share
1.8%
Effective in $0.0580Effective out $0.1684
Token share
7.9%
Effective in $0.0742Effective out $0.2091
Token share
0.5%
Effective in $0.0937Effective out $0.5584
Token share
0.3%
Effective in $0.0958Effective out $0.2783
Token share
8.7%
Effective in $0.1020Effective out $0.4987
Token share
0.6%
Effective in $0.1065Effective out $0.2798
Token share
0.1%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

137tok/s

P50, best across providers

Latency

0.31s

P50, best provider

Reasoning effortLatency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-BenchFireworks84.7%--OpenInference86.2%81.7%Cloudflare89.7%75.4%GMICloud88.3%75.8%Crusoe90.1%73.3%

+14 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.13%

Availability over the last 3 days

Last 72 hours

Availability 99.13%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.76%

Without Routing

96.99%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for DeepSeek: DeepSeek V4 Flash 0423 (Artificial Analysis and Design Arena)
Source	Benchmark	Score
Artificial Analysis	DeepSeek V4 Flash 0420 (High) Coding Index	52.0
Artificial Analysis	DeepSeek V4 Flash 0420 (High) Agentic Index	26.3
Artificial Analysis	DeepSeek V4 Flash 0420 (High) GPQA Diamond	86.7%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) HLE	30.3%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) IFBench	73.5%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) τ²-Bench Telecom	95.6%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) AA-LCR	72.0%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) τ-Bench Banking	26.2%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) GDPval-AA	25.5%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) CritPt	3.4%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) SciCode	40.2%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) Terminal-Bench Hard	38.6%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) Terminal-Bench 4.0	3.0%
Artificial Analysis	DeepSeek V4 Flash 0420 (High) Terminal-Bench 2.1	56.9%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) GPQA Diamond	71.6%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) HLE	7.8%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) IFBench	47.2%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) τ²-Bench Telecom	94.4%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) AA-LCR	41.7%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) CritPt	0.3%
Artificial Analysis	DeepSeek V4 Flash 0420 (Non-reasoning) Terminal-Bench Hard	34.1%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) Coding Index	56.2
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) Agentic Index	22.2
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) GPQA Diamond	89.4%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) HLE	34.8%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) IFBench	79.2%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) τ²-Bench Telecom	95.0%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) AA-LCR	74.3%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) τ-Bench Banking	30.9%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) GDPval-AA	27.3%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) CritPt	7.1%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) SciCode	45.3%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) Terminal-Bench Hard	35.6%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) Terminal-Bench 4.0	2.5%
Artificial Analysis	DeepSeek V4 Flash 0420 (Max) Terminal-Bench 2.1	61.8%
Design Arena	DeepSeek-V4-Flash Models Arena 3D Elo	1169
Design Arena	DeepSeek-V4-Flash Models Arena Asciiart Elo	1104
Design Arena	DeepSeek-V4-Flash Models Arena Code Categories Elo	1195
Design Arena	DeepSeek-V4-Flash Models Arena Data Visualization Elo	1124
Design Arena	DeepSeek-V4-Flash Models Arena Game Development Elo	1186
Design Arena	DeepSeek-V4-Flash Models Arena SVG Elo	1145
Design Arena	DeepSeek-V4-Flash Models Arena UI Component Elo	1155
Design Arena	DeepSeek-V4-Flash Models Arena Website Elo	1199

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

1.21Ttokens

OpenClaw

OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.

214Btokens

There are many coding agents, but this one is yours.

201Btokens

claude-mem

Memory for Claude Code

118Btokens

nanobot

Ultra-lightweight personal AI agent

91.6Btokens

Rest of OpenRouter

169B

Top 5 Apps

29.8B

Not currently available through EU in-region routing. In-region routing docs

More models from DeepSeek

DeepSeek V4.1 Flash 1017

DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision Exp.

It is suited for coding, terminal, and computer-use agents, along with long-horizon tasks that must run to completion across many steps and long-context analysis. Compressed KV caching cuts cache memory to roughly a quarter of the previous Flash generation, significantly reducing costs on agentic workloads. DeepSeek positions it as the cost-efficient tier of the V4.1 family and reports that it exceeds V4 Pro on performance, speed, and task completion time.

Text1.0M context$0.038 / $0.50

DeepSeek V4.1 Flash 1017 (batch)

Text1.0M context$0.05 / $0.15

DeepSeek Pro Latest

This model always redirects to the latest model in the DeepSeek Pro family.

Text1.0M context

DeepSeek Flash Latest

This model always redirects to the latest model in the DeepSeek Flash family.

Text1.0M context

DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision Exp.

Text1.0M context$0.0507 / $0.2028

DeepSeek V4.1 Flash (batch)

Text1.0M context$0.112 / $0.336

DeepSeek V4 Flash Vision Exp

DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of DeepSeek V4 Flash 0731 from DeepSeek, adding image understanding while matching the base model on text capabilities including agents, reasoning, and world knowledge. It is a sparse mixture-of-experts model with 13B active parameters out of 284B total.

It is suited for document and chart understanding, visual question answering, and multimodal agent workflows that interleave text and images.

Text1.0M context$0.2156 / $0.6468

DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.

Text1.0M context$0.0999 / $8

DeepSeek V4 Flash Latest

This model always redirects to the latest model in the DeepSeek V4 Flash family.

Text1.0M context

DeepSeek V4 Flash 0731

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. This is the GA release of DeepSeek V4 Flash.

Text1.0M context$0.0434 / $0.30

DeepSeek V4 Flash 0731 (batch)

Text1.0M context$0.048 / $0.144

DeepSeek V4 Pro 0423

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding, and long-horizon agent workflows, with strong performance across knowledge, math, and software engineering benchmarks.

Built on the same architecture as DeepSeek V4 Flash, it introduces a hybrid attention system for efficient long-context processing. Reasoning efforts high and xhigh are supported; xhigh maps to max reasoning. It is well suited for complex workloads such as full-codebase analysis, multi-step automation, and large-scale information synthesis, where both capability and efficiency are critical

Text1.0M context$0.2079 / $0.4157

DeepSeek V4 Flash 0423

Text1.0M context$0.014 / $0.45

DeepSeek V3.2 Speciale

DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance. It builds on DeepSeek Sparse Attention (DSA) for efficient long-context processing, then scales post-training reinforcement learning to push capability beyond the base model. Reported evaluations place Speciale ahead of GPT-5 on difficult reasoning workloads, with proficiency comparable to Gemini-3.0-Pro, while retaining strong coding and tool-use reliability. Like V3.2, it benefits from a large-scale agentic task synthesis pipeline that improves compliance and generalization in interactive environments.

Text131K context

DeepSeek V3.2

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism that reduces training and inference cost while preserving quality in long-context scenarios. A scalable reinforcement learning post-training framework further improves reasoning, with reported performance in the GPT-5 class, and the model has demonstrated gold-medal results on the 2025 IMO and IOI. V3.2 also uses a large-scale agentic task synthesis pipeline to better integrate reasoning into tool-use settings, boosting compliance and generalization in interactive environments.

Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs

Text164K context$0.2088 / $0.3096

DeepSeek V3.2 Exp

DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism designed to improve training and inference efficiency in long-context scenarios while maintaining output quality. Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs

The model was trained under conditions aligned with V3.1-Terminus to enable direct comparison. Benchmarking shows performance roughly on par with V3.1 across reasoning, coding, and agentic tool-use tasks, with minor tradeoffs and gains depending on the domain. This release focuses on validating architectural optimizations for extended context lengths rather than advancing raw task accuracy, making it primarily a research-oriented model for exploring efficient transformer designs.

Text164K context$0.27 / $0.41

DeepSeek V3.1 Terminus

DeepSeek-V3.1 Terminus is an update to DeepSeek V3.1 that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's performance in coding and search agents. It is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes. It extends the DeepSeek-V3 base with a two-phase long-context training process, reaching up to 128K tokens, and uses FP8 microscaling for efficient inference. Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs

The model improves tool use, code generation, and reasoning efficiency, achieving performance comparable to DeepSeek-R1 on difficult benchmarks while responding more quickly. It supports structured tool calling, code agents, and search agents, making it suitable for research, coding, and agentic workflows.

Text164K context$0.27 / $1

DeepSeek V3.1

DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context training process, reaching up to 128K tokens, and uses FP8 microscaling for efficient inference. Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs

It succeeds the DeepSeek V3-0324 model and performs well on a variety of tasks.

Text164K context$0.25 / $0.95

DeepSeek V3.1 Base

This is a base model, trained only for raw next-token prediction. Unlike instruct/chat models, it has not been fine-tuned to follow user instructions. Prompts need to be written more like training text or examples rather than simple requests (e.g., “Translate the following sentence…” instead of just “Translate this”).

DeepSeek-V3.1 Base is a 671B parameter open Mixture-of-Experts (MoE) language model with 37B active parameters per forward pass and a context length of 128K tokens. Trained on 14.8T tokens using FP8 mixed precision, it achieves high training efficiency and stability, with strong performance across language, reasoning, math, and coding tasks.

Text164K context

R1 Distill Qwen 7B

DeepSeek-R1-Distill-Qwen-7B is a 7 billion parameter dense language model distilled from DeepSeek-R1, leveraging reinforcement learning-enhanced reasoning data generated by DeepSeek's larger models. The distillation process transfers advanced reasoning, math, and code capabilities into a smaller, more efficient model architecture based on Qwen2.5-Math-7B. This model demonstrates strong performance across mathematical benchmarks (92.8% pass@1 on MATH-500), coding tasks (Codeforces rating 1189), and general reasoning (49.1% pass@1 on GPQA Diamond), achieving competitive accuracy relative to larger models while maintaining smaller inference costs.

Text131K context

DeepSeek: DeepSeek V4 Flash 0423

Providers