Pixtral 12B vs Llama 3.3 70B
Side-by-side API pricing comparison · Mistral vs Meta
🏆
Pixtral 12B is 590% cheaper on blended cost ($0.100 vs $0.690/Mtok)
Pixtral 12B
by Mistral
Current open weights Open weightsInput
$0.100/Mtok
Output
$0.100/Mtok
✓ Cheaper
| Blended avg | $0.100/Mtok |
|---|---|
| Context | 128K tokens |
| Modality | text, image |
| Parameters | 12B |
| Released | Nov 1, 2024 |
Llama 3.3 70B
by Meta
Current open weights Open weightsInput
$0.590/Mtok
Output
$0.790/Mtok
| Blended avg | $0.690/Mtok |
|---|---|
| Context | 128K tokens |
| Modality | text |
| Parameters | 70B |
| Released | Dec 6, 2024 |
Cost at scale — 1M tokens (50/50 input/output)
| Volume | Pixtral 12B | Llama 3.3 70B | Savings |
|---|---|---|---|
| 1M tokens | $0.1 | $0.69 | $0.59 (85.5%) |
| 10M tokens | $1 | $6.9 | $5.9 (85.5%) |
| 100M tokens | $10 | $69 | $59 (85.5%) |
| 1000M tokens | $100 | $690 | $590 (85.5%) |
Summary
Pixtral 12B by Mistral costs $0.100/Mtok input and $0.100/Mtok output, with a 128K-token context window. It supports text, image input.
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.
On a blended cost basis, Pixtral 12B is 590% cheaper than Llama 3.3 70B.
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.