DeepSeek

DeepSeek V4 Flash 0423

deepseek/deepseek-v4-flash

Model weightsReleased Apr 24, 2026MIT

ThinkingToolsopen weights

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...

Run in ChatOSS Cloud Plans

Cost / 1M tokens

Very low usage

$0.047

input

$0.0095

cached input

$0.095

output

Context Window

Up to 131K max output

Modalities

text → text

DeepSeek tokenizer

Quickstart & API

deepseek/deepseek-v4-flash

Open the model picker in ChatOSS Desktop (Code, Kanban, or Chat) and select or paste this model ID.

Best for & Strengths

Key Strengths

Fast codingOpen sourceEfficiency

Recommended Workloads in ChatOSS

Coding assistants
Fast development
Self-hosted AI

Available in ChatOSS Code, Kanban Orchestrate, and Chat modes via Plus, Max, or BYOK.

Verified Evaluations

Benchmarks & Frontier Comparison

How DeepSeek V4 Flash 0423 compares against leading open-weight and proprietary frontier models across independent agentic coding and reasoning suites.

View full Top 40 Leaderboard →

DeepSeek V4 Flash 0423DeepSeek V4.1 FlashGLM 5.3Kimi K3.1 Pro

Benchmark (Metric)	DeepSeek V4 Flash 0423	DeepSeek V4.1 Flash	GLM 5.3	Kimi K3.1 Pro	Claude Opus 5.5
Agentic & Software Engineering
ChatOSS Agentic Coding Index (%)	44.2%	63.2%	58.3%	65%	74.1%
SWE-Bench Verified (%)	88.8%	91.5%	95.4%	94.8%	97%
DeepSWE v1.1 (Pass@1 %)	54.4%	77%	69%	73.5%	74.2%
Vibe Code Bench v1.1 (%)	74.7%	87.7%	78.1%	88.4%	90.3%
Terminal-Bench 4.0 (%)	9.09%	42.3%	42%	46.2%	65.2%
Code Migration (%)	38.6%	45.6%	44.2%	52%	66.7%
Intelligence & Token Economics
Artificial Analysis Intelligence Index (0–100)	24.4	39.5	44.8	55.1	57.6
Input Price (/ 1M tokens)	$0.047	$0.15	$1.40	$2.00	$4.00
Output Price (/ 1M tokens)	$0.095	$0.60	$4.40	$10.00	$20.00

ChatOSS Agentic Coding Index

ChatOSS Benchmark Audit · 4-benchmark composite mean

44.2%

SWE-Bench Verified

Vals AI · Terminus 2

88.8%

DeepSWE v1.1

DeepSWE / Official Audit · Mini-SWE-Agent

54.4%

Vibe Code Bench v1.1

Vals.ai · OpenHands

74.7%

Terminal-Bench 4.0

Vals.ai / Artificial Analysis · Terminus 2

9.09%

Terminal-Bench 2.1

Vals.ai · Terminus 2

67.04%

Code Migration

Vals.ai · Vals evaluation harness

38.6%

Artificial Analysis Intelligence Index

Artificial Analysis

24.4

Design Arena: 3D

Design Arena · Rank #56 (49.3% win)

1186 Elo

Design Arena: ASCIIART

Design Arena · Rank #59 (42.7% win)

1107 Elo

Design Arena: Code Generation

Design Arena · Rank #62 (48.9% win)

1204 Elo

Design Arena: Data Visualization

Design Arena · Rank #93 (40.7% win)

1127 Elo

Active Provider Endpoints (17)

Upstream inference providers serving DeepSeek V4 Flash 0423, with per-provider pricing, context limits, quantization, and 24-hour uptime.

Provider	Input / 1M	Output / 1M	Cache Read / 1M	Context	Max Output	Quantization	Uptime (24h)
Relacecache relace/fp4	$0.045	$0.45	$0.045	1M	944K	fp4	100.0%
Baiducache baidu/fp8	$0.047	$0.095	$0.0095	1M	131K	fp8	99.9%
StreamLakecache streamlake/fp8	$0.07	$0.14	$0.014	1M	384K	fp8	100.0%
DeepInfracache deepinfra/fp8	$0.09	$0.18	$0.018	1M	66K	fp8	100.0%
GMICloudcache gmicloud/fp8	$0.091	$0.182	$0.018	1M	944K	fp8	100.0%
Venicecache venice	$0.097	$0.193	$0.02	1M	33K	—	100.0%
DigitalOceancache digitalocean	$0.098	$0.196	$0.02	1M	384K	—	100.0%
Fireworkscache fireworks	$0.105	$0.21	$0.021	1M	944K	—	73.7%
SiliconFlowcache siliconflow/fp8	$0.13	$0.28	$0.028	1M	393K	fp8	97.9%
Crusoecache crusoe/fp8	$0.14	$0.28	$0.03	1M	944K	fp8	100.0%
AtlasCloudcache atlas-cloud/fp4	$0.14	$0.28	$0.028	1M	393K	fp4	99.9%
Parasailcache parasail/fp8	$0.14	$0.28	$0.07	1M	944K	fp8	100.0%
Alibabacache alibaba/fp8/us-east-1	$0.15	$0.30	$0.03	1M	393K	fp8	90.2%
Mancer 2 mancer/fp8	$0.19	$0.50	—	1M	944K	fp8	99.1%
Azurecache azure/us	$0.21	$0.56	$0.031	1M	384K	—	89.2%
Cloudflarecache cloudflare	$0.44	$1.32	$0.014	384K	346K	—	100.0%
Wafercache wafer/mxfp4	$0.60	$0.17	$0.20	1M	944K	mxfp4	100.0%

Supported API Parameters

Parameters accepted by this model when called through ChatOSS or OpenRouter.

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_completion_tokensmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_atop_ktop_logprobstop_p

Compare Similar Models

Browse all models →

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...

$0.15 / $0.60Coding 63.2%

DeepSeek

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...

$0.14 / $0.42Coding 63.2%

DeepSeek

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,...

$0.216 / $0.647Intel 34.8

DeepSeek

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.

$0.66 / $1.98Coding 46.9%

Anthropic

Claude Opus 5.5

Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...

$4.00 / $20.00Coding 74.1%

Moonshot AI

Kimi K3.1 Pro

Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...

$2.00 / $10.00Coding 65%