VerdictPal
Join

Models / DeepSeek

DeepSeek V4 Flash Vision (Reasoning, Max Effort)

DeepSeek V4 · Released 2026-08-21

AA Intelligence Index 51.5. Listed on Artificial Analysis with API pricing $0.44/$1.32 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#24 of 166 on AA Index · current snapshot
52 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite59
ContextUnlisted
Input / 1M$0.44
Output / 1M$1.3
Knowledge cutoffunpublished
Output speed111 t/s
TTFT (default API)0.61s
Statussolid

Family profile

Best published score in each covered benchmark family.

73/100 avg
Reasoning
91
Coding
47
Agentic
74
Long context
78

4 tested benchmark families

Benchmark placements

Where DeepSeek V4 Flash Vision (Reasoning, Max Effort) places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
τ³-Bench BankingAgentic#15/ 10641Artificial Analysis2026-09-01
AA-LCRLong context#19/ 17978Artificial Analysis2026-09-01
GPQA DiamondReasoning#28/ 18591Artificial Analysis2026-09-01
AA output speedPerformance#33/ 185Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#34/ 18452Artificial Analysis2026-09-01
Terminal-BenchAgentic#38/ 18474Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#54/ 18035Artificial Analysis2026-09-01
SciCodeCoding#69/ 18047Artificial Analysis2026-09-01
AA time to first tokenPerformance#74/ 185Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where DeepSeek V4 Flash Vision (Reasoning, Max Effort) shows up, it's highlighted.

ReasoningDeepSeek V4 Flash Vision (Reasoning, Max Effort) · 91
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingDeepSeek V4 Flash Vision (Reasoning, Max Effort) · 47
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticDeepSeek V4 Flash Vision (Reasoning, Max Effort) · 74
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextDeepSeek V4 Flash Vision (Reasoning, Max Effort) · 78
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for DeepSeek V4 Flash Vision (Reasoning, Max Effort) in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context78%Artificial Analysis2
AA output speedPerformance111 t/sArtificial Analysis2
AA time to first tokenPerformance0.61sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning51.5%Artificial Analysis2
GPQA DiamondReasoning91.3%Artificial Analysis2
Humanity's Last ExamReasoning34.5%Artificial Analysis2
SciCodeCoding46.6%Artificial Analysis2
τ³-Bench BankingAgentic41.03%Artificial Analysis2
Terminal-BenchAgentic74.16%Artificial Analysis2
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning91
Coding47
Agentic74
Long context78
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for DeepSeek V4 Flash Vision (Reasoning, Max Effort). Scores link to the original source; gaps mean no public row exists yet.

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for DeepSeek V4 Flash Vision (Reasoning, Max Effort).

Reasoning

Coding

Agentic

Long context

Performance

Sources