DeepSeek V4 Flash 0420
Apr 2026
Older versionReplace withDeepSeek V4.1 Flash 1017Newer version
GPT-6 LunaCheaper, as good
MiMo v2.6 ProOpen weights
DeepSeek V4.1 Flash 1017Newer versionGPT-6 LunaCheaper, as goodMiMo v2.6 ProOpen weightsDeepSeek V4 Flash 0420Compare inon
Benchmarks of DeepSeek V4 Flash 042023
BenchmarkByScoreRank
Overall1
Knowledge4
Math1
Science3
Coding4
Agents5
Health & law1
Documents2
Legacy · saturated or no longer reported2
No benchmark matches
Rankingat high effort?
Retired models are left out of the current rankings.
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT-6 Lunamax+56%-34%$0.068Function calling only with none · Chat Completions+3OpenMiMo v2.6 Prothinking+90%+30%$0.13Pass back reasoning_content+3CheaperMiMo v2.6 FlashthinkingOpen+55%-39%$0.062Pass back reasoning_content+3
CheaperClaude Haiku 5.5high+55%-23%$0.079Remove sampling parameters+5CheaperGPT-5.6 Lunaxhigh+42%-17%$0.085ModelScoreCostPer taskWhat changes
GPT-6 Lunamax+56%-34%$0.068Function calling only with none · Chat Completions+3MiMo v2.6 FlashthinkingOpen+55%-39%$0.062Pass back reasoning_content+3Claude Haiku 5.5high+55%-23%$0.079Remove sampling parameters+5GPT-5.6 Lunaxhigh+42%-17%$0.085MiMo v2.5 ProthinkingOpen+7%-47%$0.054ModelScoreCostPer taskWhat changes
GPT-6 Lunamedium+23%-83%$0.017Function calling only with none · Chat Completions+3Claude Haiku 5.5low+21%-76%$0.024Remove sampling parameters+5GPT-5.6 Lunahigh+32%-57%$0.044MiMo v2.6 FlashthinkingOpen+55%-39%$0.062Pass back reasoning_content+3GPT-6.1 Sollow+73%+27%$0.13Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen+90%+30%$0.13Pass back reasoning_content+3DeepSeek V4.1 Flash 1017maxOpen+85%+155%$0.26Pass back reasoning_content · Chat Completions+4
GLM 5.3maxOpen+84%+1,855%$2.01Reasoning cannot be disabled+3
Step 5 PreviewthinkingOpen+79%+901%$1.03web_search not available · Chat Completions+2
Kimi K3maxOpen+79%+1,850%$2.00Sampling parameters are fixed+3ModelScoreCostPer taskWhat changes
DeepSeek V4.1 Flash 1017maxOpen+85%+155%$0.26Pass back reasoning_content · Chat Completions+4DeepSeek V4 Pro 0813non-thinkingOpen-17%+247%$0.36Pass back reasoning_content · Chat Completions+4DeepSeek V3.2 SpecialethinkingOpen-41%——Context: 1M → 128KDeepSeek V3.1 Terminusnon-thinkingOpen-43%—$0.27 / $0.95Context: 1M → 128KDeepSeek V3.1non-thinkingOpen-44%—$0.25 / $0.95Context: 1M → 128KR1 0528thinkingOpen-46%—$0.39 / $1.63Context: 1M → 128KDeepSeek V3 0324non-thinkingOpen-60%-79%$0.021Context: 1M → 128KDeepSeek V3non-thinkingOpen-65%-81%$0.02Context: 1M → 128KAll DeepSeek models →Run it yourself
Fits onhardware needs not verified yet
Licencenot verified
History
4 versionsBenchmarks
all categoriesChat & Knowledge
AA Omniscience Index-23.1
MLCR-AA6.7
GPQA Diamond86.7
Coding
SciCode40.2
Agents & Tool Use
τ³-Banking · AA26.2
AA-Briefcase 1.1 · Elo1,043.6
τ²-bench · AA95.6
Writing & Language
FAQ
How much does DeepSeek V4 Flash 0420 cost?›
$0.094 per 1M input tokens and $0.19 per 1M output tokens. A typical benchmark task costs about $0.10 at high effort.
Is DeepSeek V4 Flash 0420 free to use?›
There is no free API; the cheapest paid price is $0.094 per 1M input tokens. The weights are open, so you can run it yourself.
How can I use DeepSeek V4 Flash 0420?›
Through the DeepSeek API.
What is the context window of DeepSeek V4 Flash 0420?›
1M tokens.
When was DeepSeek V4 Flash 0420 released? Is there a newer version?›
Released 2026-04-24. It has been superseded by DeepSeek V4 Flash 0731.
Is DeepSeek V4 Flash 0420 open source?›
Yes, the weights are open.
Can I run DeepSeek V4 Flash 0420 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is DeepSeek V4 Flash 0420?›
DeepSeek V4 Flash 0420 is an open-weights language model from DeepSeek, released Apr 2026.
What are the best alternatives to DeepSeek V4 Flash 0420?›
GPT-6 Luna (cheaper), MiMo v2.6 Pro (open weights).