Last updated: April 5, 2026 · Reviewed by Daniel Ashford

💰 DeepSeek V3 — Review & Scores

by DeepSeek · Rank #9 · Value

Near-frontier quality at 1/30th the cost. Open weights available.

LLM JUDGE INDEX™
85.5
+4.2/week

Evaluation Scores

🎯 Accuracy87
🧠 Reasoning86
🛡️ Safety82
💻 Coding90
Creativity83
📋 Instruction85

Specifications & Pricing

Input
$0.55/M
Output
$2.19/M
Context
128K
Latency
2s
Arena
#10
Index
85.5
DA
Daniel Ashford
Founder & Lead Evaluator · 200+ models evaluated
Try DeepSeek →
🏆 Certified
Best Value Q2 2026
Best For
Budget apps
High volume
Coding tasks
Compare
vs Claude Opus 4
vs GPT-5.3 Codex
vs Gemini 2.5 Ultra
vs Claude Sonnet 4
SPONSORED

Evaluate DeepSeek V3 on your production data.

Try Evidently Free →