AA-OmniscienceVerified
49.1%
AA-Omniscience Accuracy
Models / DeepSeek
DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
DeepSeek V4 · Released 2026-08-13
AA Intelligence Index 53.2. Listed on Artificial Analysis with API pricing $1.32/$3.96 per 1M tokens after the 17 Aug 2026 list-price change. 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.
#17 of 168 on AA Index · current snapshot
53 AA Index · 10 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.Composite60
Context128K
Input / 1M$1.3
Output / 1M$4.0
Knowledge cutoffopen
Output speed51 t/s
TTFT (default API)0.98s
Statussolid
Family profile
Best published score in each covered benchmark family.
69/100 avg
- Knowledge
- 49
- Reasoning
- 93
- Coding
- 49
- Agentic
- 79
- Long context
- 75
5 tested benchmark families
Benchmark placements
Per-benchmark positions
Where DeepSeek V4 Pro 0813 (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.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | Reasoning | #14/ 185 | 93 | Artificial Analysis | 2026-09-01 |
| τ³-Bench Banking | Agentic | #17/ 106 | 40 | Artificial Analysis | 2026-09-01 |
| Artificial Analysis Intelligence Index | Reasoning | #22/ 187 | 53 | Artificial Analysis | 2026-09-01 |
| Terminal-Bench | Agentic | #29/ 184 | 79 | Artificial Analysis | 2026-09-01 |
| Humanity's Last Exam | Reasoning | #30/ 180 | 41 | Artificial Analysis | 2026-09-01 |
| AA-LCR | Long context | #45/ 179 | 75 | Artificial Analysis | 2026-09-01 |
| SciCode | Coding | #50/ 180 | 49 | Artificial Analysis | 2026-09-01 |
| AA time to first token | Performance | #64/ 186 | Artificial Analysis | 2026-09-01 | |
| AA output speed | Performance | #65/ 187 | Artificial Analysis | 2026-09-01 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where DeepSeek V4 Pro 0813 (Reasoning, Max Effort) shows up, it's highlighted.
KnowledgeDeepSeek V4 Pro 0813 (Reasoning, Max Effort) · 49
- 191
- 291
- 391
ReasoningDeepSeek V4 Pro 0813 (Reasoning, Max Effort) · 93
- 195
- 295
- 395
Math
- 199
- 299
- 399
CodingDeepSeek V4 Pro 0813 (Reasoning, Max Effort) · 49
- 196
- 296
- 396
AgenticDeepSeek V4 Pro 0813 (Reasoning, Max Effort) · 79
- 1100
- 2100
- 3100
Long contextDeepSeek V4 Pro 0813 (Reasoning, Max Effort) · 75
- 183
- 283
- 381
Tool use
- 189
- 287
- 387
Safety
- 192
- 292
- 391
Human preference
- 195
- 292
- 391
Newest receipts
10 recentLast 120 days
Benchmark rows added to the public ledger for DeepSeek V4 Pro 0813 (Reasoning, Max Effort) in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA-Omniscience | Knowledge | 49.1% | Artificial Analysis — AA-Omniscience | 5 |
| AA-LCR | Long context | 75.33% | Artificial Analysis | 5 |
| AA output speed | Performance | 51 t/s | Artificial Analysis | 5 |
| AA time to first token | Performance | 0.98s | Artificial Analysis | 5 |
| Artificial Analysis Intelligence Index | Reasoning | 53.2% | Artificial Analysis | 5 |
| GPQA Diamond | Reasoning | 92.8% | Artificial Analysis | 5 |
| Humanity's Last Exam | Reasoning | 41% | Artificial Analysis | 5 |
| SciCode | Coding | 49.2% | Artificial Analysis | 5 |
| τ³-Bench Banking | Agentic | 39.59% | Artificial Analysis | 5 |
| Terminal-Bench | Agentic | 78.65% | Artificial Analysis | 5 |
Family coverage
Best score per family
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.
Price vs. performance
Where DeepSeek V4 Pro 0813 (Reasoning, Max Effort) sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA
Full benchmark ledger
Every catalog benchmark for DeepSeek V4 Pro 0813 (Reasoning, Max Effort). Scores link to the original source; gaps mean no public row exists yet.
11 of 37 catalog benchmarks have a sourced row for DeepSeek V4 Pro 0813 (Reasoning, Max Effort).
Knowledge
Reasoning
53.2%
AA Intelligence Index v4.1
GPQA DiamondVerified
92.8%
GPQA Diamond (AA run)
Humanity's Last ExamVerified
41%
HLE (AA run)
Coding
SciCodeVerified
49.2%
SciCode (AA run)
Agentic
Terminal-BenchVerified
78.65%
Terminal-Bench v2.1 (AA run)
Vals IndexVerified
56.23%
Vals Index
τ³-Bench BankingVerified
39.59%
τ³-Bench Banking (AA run)
Long context
AA-LCRVerified
75.33%
AA-LCR (AA run)
Performance
AA output speedVerified
51 t/s
Median output tokens/s (1k prompt, default provider)
AA time to first tokenVerified
0.98s
Median time to first token (1k prompt, default provider)
Sources
Where the numbers come from
- Artificial Analysis DeepSeek V4 Pro· checked 2026-08-17