Head-to-Head API Cost Benchmark

Mistral Large 2 vs Mistral Small 3: API Price & Latency Comparison

On a standard B2B production workload (5,000 req/day), Mistral Small 3 is 95.0% cheaper ($46.12/mo vs $922.50/mo), while Mistral Small 3 leads in token generation speed (150 tok/s).

Side-by-Side Specification & Billing Matrix

Metric Mistral Large 2 (Mistral AI (La Plateforme)) Mistral Small 3 (Mistral AI (La Plateforme)) Advantage
Standard Input / 1M Tokens $2.000 $0.100 Mistral Small 3
Cached Input / 1M Tokens $1.000 $0.050 Mistral Small 3
Output Generation / 1M Tokens $6.00 $0.30 Mistral Small 3
Intelligence / Benchmark Quality 1270 LMSYS Elo 1215 LMSYS Elo Mistral Large 2
Monthly Spend (5k req/day, 50% cache) $922.50 / mo $46.12 / mo Mistral Small 3 (-95.0%)
Context Window 128,000 tokens 32,000 tokens Mistral Large 2
Output Speed (Throughput) 90 tokens/sec 150 tokens/sec Mistral Small 3
Batch API Discount -50% (Mistral AI (La Plateforme)) -50% (Mistral AI (La Plateforme)) Mistral AI (La Plateforme)
Primary Workload Fit Multilingual European enterprise workflows (DE/FR/ES/CS/PL) with EU data residency GDPR-strict European customer support automation & structured extraction Workload Dependent
Deploy / Console Access Try Mistral Large 2 → Try Mistral Small 3 → Open Full Simulator