Models
Price comparison

DeepSeek V4 Pro

DeepSeek V4 Pro by DeepSeek: 14 offers. From €0.59 per 1M input tokens, with EU processing from €1.56.

Calculate costs for this modelCompare offersView price history

Profile

Profile
DeveloperDeepSeek
TypeLanguage models
OriginCN
Input typestext
Context (tokens)1,000,000
Max. output (tokens)384,000
Reasoningohne
Parameters (bn)1,598.84
Open weights✓ yes
LicenceMIT
VRAM at 4-bit (GB)884 (Rechenzentrum)
Released–

searchit Entity Extraction (24 Sept 2026): 63.8 %

Offers with processing in the EU (3)

ProviderCountryVariantInputOutputSourceAs of
cortecsATstandard€1.53€3.05api.cortecs.ai14 Oct 2026
LyceumDEstandard€1.56€3.13lyceum.technology13 Oct 2026
HostYourAINLstandard€2.01€4.03hostyourai.com14 Oct 2026

€ per 1M tokens. Source: providers' pricing pages, read and checked by searchit. As of 14 Oct 2026.

Offers outside the EU (11)

ProviderCountryVariantInputOutputSourceAs of
DeepSeekCNoff-peak€0.59€1.77api-docs.deepseek.com2 Oct 2026
DeepInfraUSstandard€1.16€2.32deepinfra.com14 Oct 2026
SiliconFlowSGstandard€1.34€2.80www.siliconflow.com2 Oct 2026
Novita AIUSstandard€1.43€2.86novita.ai13 Oct 2026
Venice.aiUSstandard€1.47€2.95docs.venice.ai14 Oct 2026
CrusoeUSstandard€1.56€3.11www.crusoe.ai13 Oct 2026
ParasailUSstandard€1.56€3.11www.parasail.io2 Oct 2026
ParasailUSserverless€1.56€3.11www.parasail.io2 Oct 2026
Nebius Token FactoryNLstandard€1.56€3.13tokenfactory.nebius.com14 Oct 2026
DeepSeekCNpeak€1.18€3.54api-docs.deepseek.com2 Oct 2026
io.netUSstandard€1.25€4.05api.intelligence.io.solutions14 Oct 2026

€ per 1M tokens. Source: providers' pricing pages, read and checked by searchit. As of 14 Oct 2026.

AI Act

Whether the developer has signed the Code of Practice for general-purpose AI models: unknown. What this means is explained under “EU and AI Act”. EU processing and AI Act

Benchmarks

Benchmarks
GPQA Diamond (Artificial Analysis)90.5
Humanity's Last Exam (Artificial Analysis)35.2
OTIS Mock AIME 2024-2025 (Epoch AI)95.56
FrontierMath Tiers 1-3 v2 (Epoch AI)45.3
SimpleQA Verified (Epoch AI)47
SWE-bench Verified (Epoch AI)77.6
Vectara HHEM-2.38.6
Vectara HHEM-2.3 – antwortrate97.2
Vectara HHEM-2.3 – halluzination8.6
LiveBench (Aufgabensatz 2026-01-08)73.6
searchit Extraktions-Messlauf REPO-9 (PROMPT-CRAWLER-TEXT-START, 8 Dokumente × 3 Läufe) – treffer beziehungen52.4
searchit Extraktions-Messlauf REPO-9 (PROMPT-CRAWLER-TEXT-START, 8 Dokumente × 3 Läufe) – treffer entitaeten75.3

Report an error on this page