Moderate125K contextMistral AI

Mistral Large

Mistral Large is a practical choice when you need solid reasoning and clean JSON without jumping to premium pricing. At $2 input and $6 output per 1M tokens, it sits in the moderate tier, but many real tasks still come out cheap: 100 short chats cost about $0.28, a long PDF plus questions is around $0.19, and even a 50-step agent workflow is only $0.25. The non-obvious part: it feels more cost-efficient in structured workflows than in heavy coding volume, where 1,000 completions hit $2.80.

Best for

  • Building agents that call tools and return predictable structured output.
  • Reading long documents and asking follow-up questions without cost anxiety.
  • General professional chat where you want reasoning quality above bargain models.

Not ideal for

  • Ultra-cheap high-volume coding runs where every output token cost matters.
  • Anyone chasing the latest cheaper Mistral variant instead of this older-priced family entry.

What it costs in real life

Computed from OpenRouter API pricing ($2.00 input / $6.00 output per 1M tokens)

100 short chats(50K in / 30K out)
$0.28Cheap
1 long PDF + questions(80K in / 5K out)
$0.19Cheap
1,000 coding completions(200K in / 400K out)
$2.80Moderate
Agent workflow (50 steps)(50K in / 25K out)
$0.25Cheap

Variants

NameContextInput/1MOutput/1M
Mistral: Mistral Large 3 2512256K$0.50$1.50
Mistral Large 2411128K$2.00$6.00
Mistral Large 2407128K$2.00$6.00
Mistral Large125K$2.00$6.00

Frequently Asked Questions

Is Mistral Large worth it for everyday work?

Yes, if your work depends on reasoning, tool calling, and structured outputs more than raw bargain pricing. It lands in the moderate tier, but common tasks stay cheap enough that the day-to-day cost usually feels lower than the headline token rates suggest.

How much does Mistral Large actually cost in the API?

The listed API price is $2.00 per 1M input tokens and $6.00 per 1M output tokens. In practical terms, 100 short chats cost about $0.28, while a 50-step agent workflow is around $0.25.

Should I use Mistral Large for coding or document analysis?

For document analysis, yes—it is a very sensible fit because long PDF plus Q&A usage is cheap at about $0.19 per run. For coding, it can still work well, but high-volume completion workloads are less compelling since 1,000 coding completions come to $2.80.

Capabilities

Vision
Tool calling
Structured output
Reasoning
Open weights
Long context

Cheapest access path

If you already pay for Le Chat Pro or Le Chat Team, check that first before adding another subscription through StackTrim AI. On API pricing, short chats and document Q&A are inexpensive in practice, with 100 short chats at $0.28 and a long PDF workflow at $0.19.

Included in 2 subscriptions

reasoningstructured outputtool uselong contextmoderate costagent workflows