Meta open architecture with 327,680 tokens supported on Together AI serverless API endpoint (base architecture supports up to 1M context on dedicated clusters) Need a token count first? Use the Llama token counter.
meta-llama/Llama-4-Scout-17B-16E-Instruct| Scenario | Input Tokens | Output Tokens | Uncached Cost | With Prompt Caching |
|---|---|---|---|---|
| Short Chat Query | 1,000 | 500 | $0.00047 | N/A |
| Document Summarization | 10,000 | 2,000 | $0.00298 | N/A |
| Codebase & Context Analysis | 100,000 | 20,000 | $0.0298 | N/A |
| Batch Corpus Processing | 1,000,000 | 100,000 | $0.2390 | N/A |
Tier 1 (free tier) reference limits — your account's actual quota may be higher.
Ideal for production workloads demanding balanced capabilities, deep context depth (328k tokens), and reliability from Meta / Together. Excellent when predictable tokenomics and prompt caching support are paramount.
If your use-case requires sub-second streaming latency or ultra-high frequency classification at micro-cent pricing, consider lighter budget options such as Gemini Flash-Lite or Claude Haiku. For deep formal logic, consider dedicated reasoning models like o3.
For Llama 4 Scout (17B Active / 109B MoE) — Together AI, 1 million input tokens costs $0.18, while 1 million output tokens costs $0.59. If using prompt caching, repetitive input prefixes are discounted to $0.05 per million.
Llama 4 Scout (17B Active / 109B MoE) — Together AI features a maximum context window of 327,680 tokens (~245,760 words), with a maximum output limit of 16,384 tokens per completion.
Llama 4 Scout (17B Active / 109B MoE) — Together AI utilizes the Meta Llama 3/4 Tiktoken (128k vocabulary). Token counting on TokenMath runs client-side to ensure maximum privacy.
No, Llama 4 Scout (17B Active / 109B MoE) — Together AI does not currently advertise prompt caching discounts on its standard API tier.