DeepSeek-V4-Flash-Vision-Exp
DeepSeek-V4-Flash-Vision-Exp is behind the leaders in coding, instruction following, and math. Too few results yet to rate reasoning, agentic tasks, safety, long context, multimodal tasks, multilingual tasks, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Reasoning, Agentic, Safety, Long Context, Multimodal, Multilingual or Factuality.
Price
$0.22input$0.65outputper million tokens
From deepseek · 3 providers tracked · All prices
Evidence
16results on16benchmarks
- 7 independently verified
- 9 vendor-reported
From 3 sources · latest Oct 8, 2026 · How verification works
API features
Tool callingStructured outputsReasoning
As listed by OpenRouter
DeepSeek-V4-Flash-Vision-Exp benchmark results
16 results on 16 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
33.6% behind the leader3 of 10 ranked benchmarks measured
- 83.90Oct 7, 2026
- LiveBench · Agentic Coding65.10Oct 8, 2026livebench_agentic_coding@2026-06-25
- LiveBench · Coding68.20Oct 8, 2026livebench_coding@2026-06-25
37.7% behind the leader1 of 3 ranked benchmarks measured
- LiveBench · Instruction Following70.96Oct 8, 2026livebench_instruction_following@2026-06-25
39.4% behind the leader1 of 5 ranked benchmarks measured
- LiveBench · Mathematics87.81Oct 8, 2026livebench_math@2026-06-25
0 of 6 ranked benchmarks measured
- LiveBench · Reasoning85.40Oct 8, 2026livebench_reasoning@2026-06-25
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- livebench_language80.36Oct 8, 2026livebench_language@2026-06-25
- livebench_data_analysis79.48Oct 8, 2026livebench_data_analysis@2026-06-25
- 35.00Oct 7, 2026
- 27.30Oct 7, 2026
- 63.60Oct 7, 2026
- 59.30Oct 7, 2026
Show 4 more resultsHide 4 results
- AutomationBench Public25.70Aug 31, 2026AutomationBench (Public)
- 75.30Aug 31, 2026
- 57.70Oct 7, 2026
- 75.90Aug 31, 2026
DeepSeek-V4-Flash-Vision-Exp: common questions
Who makes DeepSeek-V4-Flash-Vision-Exp?
DeepSeek-V4-Flash-Vision-Exp is made by DeepSeek.
When was DeepSeek-V4-Flash-Vision-Exp released?
DeepSeek-V4-Flash-Vision-Exp's weights were first published on Hugging Face on Aug 31, 2026.
What is DeepSeek-V4-Flash-Vision-Exp good at?
DeepSeek-V4-Flash-Vision-Exp is behind the leaders in coding, instruction following, and math. Too few results yet to rate reasoning, agentic tasks, safety, long context, multimodal tasks, multilingual tasks, or factuality.
How much does DeepSeek-V4-Flash-Vision-Exp cost?
DeepSeek-V4-Flash-Vision-Exp costs $0.22 per million input tokens and $0.65 per million output tokens, according to deepseek. We track its price at 3 providers. At a mix of three input tokens to one output token, it is cheaper than 70% of the 331 priced models we track.
How many benchmarks has DeepSeek-V4-Flash-Vision-Exp been tested on?
We track 16 results for DeepSeek-V4-Flash-Vision-Exp on 16 benchmarks from 3 sources, 7 of them independently verified. The latest was recorded on Oct 8, 2026.
Which API features does DeepSeek-V4-Flash-Vision-Exp support?
OpenRouter lists tool calling, structured outputs, and reasoning for DeepSeek-V4-Flash-Vision-Exp.
About this record
Where DeepSeek-V4-Flash-Vision-Exp's numbers come from, and every name it appears under.
- Tracked since
- Aug 21, 2026
- Newest source mention
- Sep 11, 2026
Where the results come from
Verification: 16 scores · 7 independently verified · 9 vendor-reported. How these tiers are assigned
From 3 sources on 3 sites. livebench.ai supplies 7 of them; the 7 independently verified results come from 1 site. Bars are coloured by trust tier.
- livebench.ai7
- api.llm-stats.com6
- huggingface.co3