Vascend
HorizonVascend IDEVascend CLICommunityBenchmarkContact
Sign inJoin Discord
HomeHorizonIDECLISign in
ChangelogIntelligenceCost per taskCoding Agent IndexCyber IndexDecision IndexBenchmarksSpeedLatencyPriceLeaderboard
Benchmark

DeepSeek V4 Flash Vision (Max)

  • DeepSeek
  • Release · August 21, 2026
  • Proprietary
  • Reasoning

Intelligence

Intelligence Index

#57 of 266
  1. 37.938MiMo-V2.6-Flash
  2. 37.838Claude Haiku 5.5 (High)
  3. 3636DeepSeek V4 Pro 0813 (Max)
  4. 35.936Claude Sonnet 5.5 (Low)
  5. 35.235Agnes 2.5 Pro Beta
  6. 34.835DeepSeek V4 Flash Vision (Max)
  7. 34.635GPT-6 Luna (Xhigh)
  8. 34.535Claude Haiku 5.5 (Medium)
  9. 34.334GLM-5.3 (Low)
  10. 34.234GPT-5.6 Terra (High)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy38.6%#58
  • Humanity's Last Exam34.5%#68
  • Terminal-Bench 4.012.1%#57
  • AA-LCR81.3%#34
  • SciCode49.7%#64
  • CritPt10.9%#66
  • MMMU-Pro74.8%#50
  • GPQA Diamond91.3%#20
  • τ²-Bench Banking41%#15
DeepSeek V4 Flash Vision (Max)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #32 of 162
  1. 183183Nova 2.0 Lite (Non-reasoning)
  2. 181181Claude Haiku 5.5 (High)
  3. 180180Nova 2.0 Lite (Medium)
  4. 179179Claude Haiku 5.5 (Medium)
  5. 178178Mistral Small 4 (Reasoning)
  6. 177177DeepSeek V4 Flash Vision (Max)
  7. 173173Mistral Small 4 (Non-reasoning)
  8. 173173Claude Haiku 5.5 (Low)
  9. 173173gpt-oss-20b (High)
  10. 173173Nova 2.0 Lite (Low)
Data from Artificial Analysis

Latency

seconds to first token · #61 of 162
  1. 1.41.4GPT-5.6 Terra (Low)
  2. 1.41.4Agnes 2.5 Pro Beta
  3. 1.41.4Reka Flash 3
  4. 1.51.5GPT-5.6 Terra (Medium)
  5. 1.51.5Magistral Small 1.2
  6. 1.61.6DeepSeek V4 Flash Vision (Max)
  7. 1.61.6Inkling Small
  8. 1.61.6Sarvam 105B (High)
  9. 1.71.7Claude Fable 5.1 (Low)
  10. 1.81.8DeepSeek V4 Pro 0813 (Max)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #84 of 169
  1. $0.530.53Inkling Small
  2. $0.530.53Muse Glimmer (High)
  3. $0.540.54MiMo-V2.6-Pro
  4. $0.60.6Beam 1.0
  5. $0.640.64Nemotron 3 Nano Omni 30B A3B Reasoning
  6. $0.660.66DeepSeek V4 Flash Vision (Max)
  7. $0.670.67Llama 3.3 Instruct 70B
  8. $0.70.7Qwen3.7 Plus
  9. $0.750.75Magistral Small 1.2
  10. $0.840.84Qwen3.6 35B A3B (Reasoning)
Data from Artificial Analysis

Nearby models

  • MiMo-V2.6-FlashXiaomi37.9
  • Claude Haiku 5.5 (High)Anthropic37.8
  • DeepSeek V4 Pro 0813 (Max)DeepSeek36
  • Claude Sonnet 5.5 (Low)Anthropic35.9
  • Agnes 2.5 Pro BetaSapiens AI35.2
  • GPT-6 Luna (Xhigh)OpenAI34.6
  • Claude Haiku 5.5 (Medium)Anthropic34.5
  • GLM-5.3 (Low)Z AI34.3
Data from Artificial Analysis

Same lab

  • DeepSeek V4.1 Flash 1017 (Max)DeepSeek45.1
  • DeepSeek V4 Pro 0813 (Non-reasoning)DeepSeek20.4
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend