Llama 3.3 70B vs Mixtral 8x7B Instruct
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
|
by Meta
2 providers:
Meta $0.590/$0.790
Together $1.04/$1.04
|
Mby Mistral
|
|
|---|---|---|
| Overview | ||
| Status | Current open weights Open weights | Current open weights Open weights |
| Released | Dec 6, 2024 | Jul 1, 2024 |
| Pricing per million tokens | ||
| Input | $0.590/Mtok | $0.700/Mtok |
| Output | $0.790/Mtok | $0.700/Mtok |
| Blended avg | $0.690/Mtok | $0.700/Mtok |
| Specifications | ||
| Context window | 128K tokens | 32K tokens |
| Parameters | 70B | 46.7B (8x7B MoE) |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
text
|
text
|
| Benchmarks sources: Vellum LLM Leaderboard (Jun 2026), Kaggle dataset / HuggingFace Open LLM Leaderboard | ||
| Avg benchmark score | 43.6 | |
| Perf / dollar | 63.2 | |
| GPQA Diamond | 45 | — |
| SWE-Bench | 22 | — |
| HumanEval | 78 | — |
| MATH 500 | 60 | — |
| Providers | ||
| Available from |
Meta — $0.590/$0.790/Mtok
Together — $1.04/$1.04/Mtok
|
Mistral — $0.700/$0.700/Mtok
|
Cost at scale — 1M tokens (50/50 input/output)
| Volume | Llama 3.3 70B | Mixtral 8x7B Instruct | Savings |
|---|---|---|---|
| 1M tokens | $0.69 | $0.7 | $0.01 (1.4%) |
| 10M tokens | $6.9 | $7 | $0.1 (1.4%) |
| 100M tokens | $69 | $70 | $1 (1.4%) |
| 1000M tokens | $690 | $700 | $10 (1.4%) |
Summary
Llama 3.3 70B by Meta costs $0.590/Mtok input and $0.790/Mtok output, with a 128K-token context window. It supports text input and is available from 2 providers.
Mixtral 8x7B Instruct by Mistral costs $0.700/Mtok input and $0.700/Mtok output, with a 32K-token context window. It supports text input.
On a blended cost basis, Llama 3.3 70B is 1.4% cheaper than Mixtral 8x7B Instruct. It also has a larger context window.
On benchmarks, Llama 3.3 70B scores higher (43.6 vs ) on average. In terms of value, Llama 3.3 70B has better performance per dollar (63.2 vs ).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.