AI Battle > Rankings > DeepSeek V4 Pro 0813 (Reasoning, Max Effort)

DeepSeek V4 Pro 0813 (Reasoning, Max Effort) — Placements in 59 Rankings (2026)

DeepSeek V4 Pro 0813 (Reasoning, Max Effort), published by DeepSeek, appears in 59 of the 129 datasets AI Battle tracks. Its strongest published placement is #3 of 14 in Model Size: Total and Active Parameters.

Last updated: . Every placement below comes from the named public dataset; AI Battle joins them by slug and changes nothing else.

All placements for DeepSeek V4 Pro 0813 (Reasoning, Max Effort)

RankingRankValue
Model Size: Total and Active Parameters#3 of 1449.0
Artificial Analysis Openness Index: Score#3 of 1244.4
Cost per Task#3 of 110.67
Cost per Task#3 of 110.67
Artificial Analysis Openness Index: Components#4 of 126.00
𝜏³-Banking: Cost per Task#8 of 200.00
Context Window#9 of 201000000
AA-Briefcase Cost per Task#9 of 200.17
Speed#5 of 1191.2
Speed#5 of 1191.2
End-to-End Response Time#10 of 201.69
Cost to Run Artificial Analysis Intelligence Index#10 of 2097.9
Latency: Time To First Answer Token#10 of 201.69
Cost to Run Artificial Analysis Intelligence Index#10 of 2097.9
Cost per Intelligence Index Task#11 of 200.01
Cost per Intelligence Index Task#11 of 200.01
GDPval-AA v2: Cost per Task#11 of 200.00
Pricing: Cache Hit, Input, and Output#12 of 201.32
Output Speed#12 of 2091.2
Pricing: Cache Hit, Input, and Output#12 of 201.32
Output Speed#12 of 2091.2
Terminal-Bench v4.0: Cost per Task#12 of 200.00
AA-Omniscience Index: Cost Breakdown#12 of 201.65
AA-Omniscience Accuracy#12 of 200.49
AA-Briefcase Output Tokens per Task#12 of 2043237
GDPval-AA v2 Leaderboard#13 of 201493
Humanity's Last Exam: Cost per Task#13 of 200.00
Artificial Analysis Finance & Accounting Index#14 of 2041.8
AA-LCR v1.1: Cost per Task#14 of 200.13
SciCode: Cost per Task#14 of 200.00
Output Tokens per Intelligence Index Task#15 of 2016991
Time per Intelligence Index Task#15 of 208.24
Time per Intelligence Index Task#15 of 208.24
Output Tokens per Intelligence Index Task#15 of 2016991
AA-Omniscience Index#16 of 200.83
Artificial Analysis Intelligence Index by Open Weights / Proprietary#16 of 2036.3
Artificial Analysis Intelligence Index by Open Weights / Proprietary#16 of 2036.3
Artificial Analysis Intelligence Index#16 of 2036.3
AA-Omniscience Index#16 of 200.83
Artificial Analysis Intelligence Index#16 of 2036.3
AA-Briefcase Elo#17 of 201265
AA-Briefcase Elo#17 of 201265
Time per Task#17 of 2033.1
𝜏³-Banking: Time per Task#17 of 202.04
Humanity's Last Exam: Score#18 of 200.41
Terminal-Bench v4.0: Score#18 of 200.14
AA-Briefcase Tool Calls Breakdown, Avg per Task#18 of 2051.1
AA-Briefcase Rubric Pass Rate by File Type (Normalized)#18 of 200.66
GDPval-AA v2 Leaderboard#18 of 201493
GDPval-AA v2: Output Tokens per Task#18 of 2026064
GDPval-AA v2: Average Turns per Task#18 of 2038.9
Terminal-Bench v4.0: Output Tokens per Task#19 of 2069376
AA-Omniscience Index: Token Usage#19 of 201249989
𝜏³-Banking: Score#19 of 200.40
Intelligence#11 of 1136.3
Intelligence#11 of 1136.3
Humanity's Last Exam: Time per Task#20 of 205.10
Terminal-Bench v4.0: Time per Task#20 of 2025.6
AA-LCR v1.1: Time per Task#20 of 200.48

Price vs intelligence

DeepSeek V4 Pro 0813 (Reintelligence index →input price / 1M → $4.00
Each dot is a model, joined by AI Battle from two public datasets — the intelligence index ranking and the published input pricing ranking. Every published value is shown as-is, no averaging. The dashed line is the value frontier: no model above or to the right of it offers a lower input price at its intelligence level.

How to read this profile

Each row is a separate measurement in a separate dataset — the position reflects the metric of that ranking, and a higher number is not "better" for every dataset (times and prices are costs). AI Battle deliberately does not average placements into a single score: datasets differ in coverage, so a total would hide more than it shows.

Raw data

Machine-readable profile: /data/models/deepseek-v4-pro.json.

Method

Placements, joins and this page are computed by AI Battle from public datasets — see the methodology for sources, limitations and corrections.