Tier

Major

Headquarters

Hangzhou, China

Founded

2023

Flagship

DeepSeek R1 / V3

Models tracked

31

Artificial Analysis

Top intelligence

44.3

Artificial Analysis Index

Cheapest model

$0.00

blended /1M

Fastest output

118

tokens / sec

Flagship Model

DeepSeek’s top-scoring model on the Artificial Analysis Intelligence Index.

Most intelligent model

DeepSeek V4 Pro (Reasoning, Max Effort)

Released 2026-04-24

Intelligence

44.3

Coding

59.4

Math

In $/1M

$0.43

Out $/1M

$0.87

Speed t/s

71

Benchmark breakdown

GPQA Diamond
89
Humanity’s Last Exam
36
SciCode
50
τ²-Bench (agentic)
96
Terminal-Bench Hard
46
IFBench
76

Models & benchmarks

Every DeepSeek model on file — intelligence, coding, price, and speed.

ModelReleasedIntelCodingMathIn $/1MOut $/1MSpeed t/sLatency s
DeepSeek V4 Pro (Reasoning, Max Effort)2026-04-2444.359.4$0.43$0.87710.91
DeepSeek V4 Pro (Reasoning, High Effort)2026-04-2443.158.7$0.43$0.87730.96
DeepSeek V4 Flash (Reasoning, Max Effort)2026-04-2440.356.2$0.14$0.281180.90
DeepSeek V4 Flash (Reasoning, High Effort)2026-04-2437.552.0$0.14$0.2800.00
DeepSeek V3.2 (Reasoning)2025-12-0132.044.292.0$0.28$0.4200.00
DeepSeek V4 Pro (Non-reasoning)2026-04-2431.2$0.43$0.87691.04
DeepSeek V3.1 Terminus (Reasoning)2025-09-2230.443.589.7$1.64$2.7500.00
DeepSeek V4 Flash (Non-reasoning)2026-04-2428.7$0.14$0.281160.91
DeepSeek V3.2 Exp (Reasoning)2025-09-2925.487.7$0.28$0.4200.00
DeepSeek V3.2 (Non-reasoning)2025-12-0124.759.0$0.28$0.4200.00
DeepSeek V3.2 Speciale2025-12-0122.296.7$0.00$0.0000.00
DeepSeek V3.1 Terminus (Non-reasoning)2025-09-2221.453.7$0.27$1.0000.00
DeepSeek V3.2 Exp (Non-reasoning)2025-09-2921.357.7$0.28$0.4200.00
DeepSeek V3.1 (Non-reasoning)2025-08-2121.049.7$0.56$1.6800.00
Via Artificial Analysis

Trajectory

DeepSeek’s model releases over time, by intelligence.

Nov 2023 · 3.0▲ +25.7 indexApr 2026 · 28.7

House strengths

Where DeepSeek’s lineup over-indexes, averaged across its models.

Governance

DeepSeek’s license posture and openness across its lineup.

Open weight

23

of 23 models

Proprietary

0

of 23 models

Commercial-OK

17

license permits

Modalities

Text · Image

MIT · 17deepseek · 4MIT + Model License (Commercial use allowed) · 2

Via LLM Stats

Media models

Image, video, and speech models, ranked by arena Elo.

ModelTypeReleasedEloRank
Janus Protext to image2025-01712#145
© 2026 NYSGPT2525 LLC