Mistral Large 3 vs Llama 3.1 405B
Mistral vs Meta · Prices checked 12 August 2026
Both price out at $0.800 per million tokens blended, making this a capability comparison rather than a cost one.
These two swap places depending on direction of traffic: Mistral Large 3 leads on an 80/20 input-heavy blend, Llama 3.1 405B on a 20/80 output-heavy one. The 5 benchmarks they both publish do not separate them; wins are shared. Mistral Large 3 reads more in one request: 256,000 tokens of context against 128,000.
Pricing and capacity
| Metric | Mistral Large 3 | Llama 3.1 405B | Difference |
|---|---|---|---|
| Input / 1M tokens | $0.500 | $0.800 | 38% |
| Output / 1M tokens | $1.50 | $0.800 | 47% |
| Blended · 70/30calculated | $0.800 | $0.800 | level |
| Blended · 50/50calculated | $1.00 | $0.800 | 20% |
| Blended · 80/20calculated | $0.700 | $0.800 | 13% |
| Blended · 20/80calculated | $1.30 | $0.800 | 38% |
| Job cost · 100K + 10Kcalculated | $0.065 | $0.088 | 26% |
| Job cost · 1M + 100Kcalculated | $0.650 | $0.880 | 26% |
| Context window | 256,000 | 128,000 | 2.0× |
Input and output rates are as published by each provider. Blended and job-cost rows are calculated comparison metrics, not published prices — how they are derived.
Benchmarks
| Benchmark | Mistral Large 3 | Llama 3.1 405B | Leader |
|---|---|---|---|
| MMLU | 84% | 88% | Llama 3.1 405B |
| GPQA Diamond | 52% | 51% | Mistral Large 3 |
| MATH | 73% | 73% | Level |
| HumanEval | 92% | 89% | Mistral Large 3 |
| LMArena Elo | 1283 Elo | 1290 Elo | Llama 3.1 405B |
Only suites both models report are listed, so a missing score is never shown as a loss. Figures are provider-published or from public leaderboards and are not independently re-run. Where a model has no score of its own it inherits its base model’s, since the weights are the same.
What each model is for
Mistral Large 3
Mistral · Frontier
Mistral's frontier model. Strong multilingual performance and Europe-friendly hosting options.
- Multilingual
- EU sovereign hosting
- Strong code
- 128K context
Llama 3.1 405B
Meta · Large
Largest dense Llama. Used as a quality benchmark for open weights, heavy to run, often hosted via Together / DeepInfra / Cerebras.
- Top open-weights quality
- Permissive license
- Strong reasoning
Frequently asked questions
Which is cheaper, Mistral Large 3 or Llama 3.1 405B?
Both price out at $0.800 per million tokens blended, making this a capability comparison rather than a cost one. Blended cost weights input at 70% and output at 30%; price your own token split for an exact answer, since the cheaper model on a blended basis is not always cheaper for your workload.
Which performs better, Mistral Large 3 or Llama 3.1 405B?
The 5 benchmarks they both publish do not separate them; wins are shared. These are the figures each provider or public leaderboard publishes, not results we re-run, and a benchmark average can hide the specific behaviour your workload depends on.
Which has the larger context window, Mistral Large 3 or Llama 3.1 405B?
Mistral Large 3 reads more in one request: 256,000 tokens of context against 128,000. The context window caps everything in a single request: prompt, retrieved documents, conversation history and the model’s own output all share it.
Can I compare Mistral Large 3 and Llama 3.1 405B on my own numbers?
Yes. Open both in the side-by-side comparison tool to see every field at once, or use the token calculator to price your actual prompt and monthly request volume against each model. Both are free and need no sign-up.
Price it for your own workload
Related matchups
Terms used on this page
Prices and scores are recomputed on every deploy from the catalogue. Prices checked 12 August 2026.