ColinBuilds.com · Current official Mistral API pricing
Mistral Large 3
- Input / 1M
- $0.50
- Output / 1M
- $1.50
- Published Mistral context window
- 256,000 tokens
Current official Mistral API pricing for mistral-large-2512 / mistral-large-latest on the Mistral Large 3 model card · as of July 8, 2026
Compare models
Same per-1M input/output rates as above. Enter your monthly token volumes to compare bills.
Mistral Large 3
- Input: 1M × $0.50 = $0.50
- Output: 1M × $1.50 = $1.50
- Total
- $2.00
GPT-4o
- Input: 1M × $2.50 = $2.50
- Output: 1M × $10.00 = $10.00
- Total
- $12.50
Difference
$10.50
GPT-4o costs $10.50 more than Mistral Large 3.
Benchmark Matrix
Reported model metrics from published sources.
Active Parameters
41B
Reported active parameter count at inference time.
Popular comparisons
30 total featuring Mistral Large 3.
Mistral Large 3 is Mistral’s current Large-model reference: a frontier-class, open-weight, multimodal model for high-complexity work where official Mistral API pricing matters.
What this model is
Mistral’s model card describes Mistral Large 3 as a state-of-the-art, open-weight, general-purpose multimodal model with a granular Mixture-of-Experts architecture. The same model card lists mistral-large-2512 and mistral-large-latest as the API names for this model.
Mistral Large 3 is the current Large lane to compare against GPT, Claude, Gemini, DeepSeek, Qwen, and other frontier models when the question is capability first, not the cheapest possible route.
Pricing notes
The calculator on this page uses Mistral’s official model-card API rate for mistral-large-2512 / mistral-large-latest: $0.50 per 1M input tokens and $1.50 per 1M output tokens.
This is a direct Mistral API price reference from the Mistral Large 3 model card, not an OpenRouter or reseller route. If Mistral changes model aliases or pricing, recheck Mistral’s official model card and pricing pages before relying on these numbers for production planning.
Benchmarks and specs
Mistral’s model card identifies Mistral Large 3 as an open-weight, general-purpose multimodal model with a granular Mixture-of-Experts architecture. It reports 675B total parameters and 41B active parameters.
The same model card lists a 256k context window and shows the model API names as mistral-large-2512 and mistral-large-latest. This page does not list benchmark scores until they are tied to a named Mistral source and test setup.
Best fit
Mistral Large 3 is best for high-complexity API comparisons, multimodal assistant work, agentic workflows, and European-provider evaluations where official Mistral pricing is required.
It is not the cheapest Mistral route. For cost-sensitive tasks, compare against Mistral Small 4 first; use Mistral Large 3 when the job needs the stronger Large tier.