← Back to Calculator

Mistral Small 3.2 vs GPT-5.3 Codex

Detailed comparison of API pricing, technical specifications, and benchmark scores. Updated 2026-07-04.

Cost at a Glance

GPT-5.3 Codex is 54.5× more expensive than Mistral Small 3.2 — switching could save 98.2% on your monthly API bill.

At 5M input + 2M output tokens per month:

Mistral Small 3.2$0.78/movsGPT-5.3 Codex$42.50/mo

Choosing Mistral Small 3.2 could save you $41.72/mo

Price Comparison (per 1M tokens)

Input (per 1M tokens)
$0.07vs$2.50
Output (per 1M tokens)
$0.20vs$15.00
Cache Write (per 1M tokens)
vs$3.13
Cache Read (per 1M tokens)
$0.01vs$0.25
Batch Input (per 1M tokens)
vs$1.25
Batch Output (per 1M tokens)
vs$7.50

Prices are per 1 million tokens. Actual costs depend on your monthly token volume. Use the calculator →

Specifications

Context Window
128Kvs1M
Max Output Tokens
64Kvs128K
Knowledge Cutoff
2025-06vs2025-10
Modalities
Text, ImagevsText
Prompt Caching
YesvsYes
Batch API
NovsYes
Thinking / Reasoning
NovsNo

Benchmark Scores

SWE-bench Verified
vs82.0
MMLU-Pro
68.0vs83.0
HumanEval
85.0vs94.5

Benchmark scores from official provider reports. Higher is better. Missing scores indicate the provider has not published that benchmark yet.

Use Cases

Mistral Small 3.2

cheapopen-weightmultimodaleu

GPT-5.3 Codex

codingspecialized

About the Providers

Mistral Small 3.2 is developed by Mistral. GPT-5.3 Codex is developed by OpenAI. This is a cross-provider comparison between Mistral and OpenAI models. For the most up-to-date pricing, always check the official API documentation from each provider.

Compare More AI Models