Llama 3.3 70B Instruct vs Mistral Large 4
API pricing, context, and independently measured performance, side by side. Llama 3.3 70B Instruct starts 85% cheaper on input than Mistral Large 4 at list prices.
Llama 3.3 70B Instruct
Meta · released Dec 6, 2024
Best input$0.10 / MTok
Best output$0.32 / MTok
Context131K
Providers11
Fastest measured0.61s° TTFT · 83 tok/s
Open weightsToolsStructured outputPrompt caching
Mistral Large 4
Mistral · released Oct 6, 2026
Best input$0.68 / MTok
Best output$2.09 / MTok
Context524K
Providers1
Fastest measured—
VisionToolsReasoningStructured outputPrompt caching
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| Llama 3.3 70B Instruct |  DeepInfra | $0.10 | $0.32 | 1.7s° | 39° tok/s | 5/5 |
| Llama 3.3 70B Instruct |  Novita | $0.135 | $0.40 | — | — | 5/5 |
| Llama 3.3 70B Instruct |  AkashML | $0.20 | $0.52 | 2.9s° | 46° tok/s | 5/5 |
| Mistral Large 4 |  Mistral | $0.68 | $2.09 | — | — | — |
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