Llama 3.3 70B Instruct vs Pixtral Large 2502
API pricing, context, and independently measured performance, side by side. Llama 3.3 70B Instruct starts 95% cheaper on input than Pixtral Large 2502 at list prices.
Llama 3.3 70B Instruct
Meta · released Dec 6, 2024
Best input$0.10 / MTok
Best output$0.32 / MTok
Context131K
Providers7
Fastest measured0.6s° TTFT · 94 tok/s
Open weightsToolsStructured outputPrompt caching
Pixtral Large 2502
Mistral · released Nov 20, 2026
Best input$2.00 / MTok
Best output$6.00 / MTok
Context128K
Providers1
Fastest measured—
Vision
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| Llama 3.3 70B Instruct | $0.10 | $0.32 | 1.7s° | 29° tok/s | 5/5 | |
| Llama 3.3 70B Instruct | $0.20 | $0.52 | 1.6s° | 57° tok/s | 5/5 | |
| Llama 3.3 70B Instruct | $0.22 | $0.50 | 2.3s° | 62° tok/s | 4/5 | |
| Pixtral Large 2502 | $2.00 | $6.00 | — | — | — |
TTFT/TPS: ModelIndex-measured medians (P50, trailing 21d) · chat10k profile (~10K in / 200 out) · ° = measured via OpenRouter routing (adds a hop) · SMOKE: of 5 programmatic capability probes passed · methodology v1.1 — full details on each model page