← Back to dashboard
DeepSeek·LLM — weighted competency·Tue 25 Aug 2026

DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 scores 87/100 competency for LLM at $1.98.

Moderate confidence
Competency
87/100
category rank #13
Cost
$1.98/1M tok
blended
Speed
49/100
Slow
Accuracy
no data yet

Score history.

Weighted LLM competency · last 12 weeks · weekly snapshot.

87/100
-1 pts vs prior
82
16 Aug19 Aug22 Aug25 Aug

What it's actually good at.

Score on each business-task subset that goes into the weighted competency.

Customer Service20% of category weight
77
Content Writing15% of category weight
81
Research & Analysis16% of category weight
84
Summarization12% of category weight
88
Translation7% of category weight
92
Reasoning16% of category weight
91
Knowledge14% of category weight
67

Pricing.

Published usage pricing. Figures refresh daily from our pricing sources.

$1.98

Usage pricing

$1.98/1M tok
Input
$1.32 / 1M tokens
Output
$3.96 / 1M tokens

Where it sits.

Position against the rest of the category. Ranked by weighted competency.

RankModelScoreCostSpeedΔ vs DeepSeek
04Claude Fable 595$20.0033 / 100+8
05Gemini 3.7 Flash94$1.5060 / 100+7
06Kimi K394$6.0018 / 100+6
07Qwen3.8 2.4T A95B92$3.0024 / 100+5
08Qwen3.8 Max89$3.0017 / 100+2
09GPT-5.589$11.2540 / 100+1
10GPT-5.6 Terra88$4.5057 / 100+1
11DeepSeek V4 Pro 081387$1.9849 / 100
12DeepSeek V4 Flash 073184$0.6678 / 100-4
13Claude Sonnet 583$4.0042 / 100-5
14Gemini 3.1 Pro Preview81$4.500 / 100-7
15GLM-5.280$2.1540 / 100-7
16GPT-5.6 Luna79$0.4560 / 100-9
17Muse Spark 1.278$2.0040 / 100-9
18GPT-5.3 Codex73$4.8160 / 100-14