Compare the listed model data for Llama 4 Scout 17B (16 Experts) by Meta Llama and Llama 4 Maverick 17B FP8 (128 Experts) by Meta Llama. Benchmark and pricing records are not a controlled test and may change.
Data last updated: 2026-10-16. Verify current model details with each provider.
What the listed data says
Llama 4 Scout 17B (16 Experts): 10M tokens; Llama 4 Maverick 17B FP8 (128 Experts): 1M tokens. Llama 4 Scout 17B (16 Experts) has the larger listed context window. Llama 4 Scout 17B (16 Experts) is listed for coding, vision, multimodal input, agentic workflows, long-context tasks; Llama 4 Maverick 17B FP8 (128 Experts) is listed for coding, reasoning, vision, multimodal input, agentic workflows, long-context tasks. Shared catalog benchmark entries: MMLU: Llama 4 Scout 17B (16 Experts) 79.9%; Llama 4 Maverick 17B FP8 (128 Experts) 85.5%; HumanEval: Llama 4 Scout 17B (16 Experts) 80.1%; Llama 4 Maverick 17B FP8 (128 Experts) 88.0%; DocVQA: Llama 4 Scout 17B (16 Experts) 88.4%; Llama 4 Maverick 17B FP8 (128 Experts) 91.6%. These records are reference data, not a controlled head-to-head test; verify current pricing and specifications with each provider.
Catalog source links
Use these model and provider links to verify current specifications, pricing, and availability.
The first listed parameter quantities are 17B active / 16 experts and 17B active / 128 experts (400B+ total); the larger quantity is 1× the smaller. Total and active MoE parameters are not directly equivalent.
The catalog release dates are April 2025 and April 2025, about 0 months apart at the precision supplied.
Cost example: 1M input and 1M output tokens
A cost example for 1M input plus 1M output tokens cannot be calculated reliably because separate numeric input and output rates are not listed for both models.
Choose Llama 4 Scout 17B (16 Experts) if…
Choose Llama 4 Scout 17B (16 Experts) when your task aligns with catalog-listed uses or strengths such as Ultra-long document processing: entire legal code sets, multi-volume research, Video understanding with multi-hour temporal context, Multi-document enterprise knowledge bases, 10 million token context — largest publicly available context window as of 2025 (roughly 30 full novels).
Choose Llama 4 Maverick 17B FP8 (128 Experts) if…
Choose Llama 4 Maverick 17B FP8 (128 Experts) when your task aligns with catalog-listed uses or strengths such as Long-document processing: entire legal contracts, technical books, large codebases, Multimodal enterprise workflows: visual document analysis, chart interpretation, RAG systems needing very long context windows, 1 million token context — processes entire books or large codebases in one session.
Catalog specifications for Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts)
Specification
Llama 4 Scout 17B (16 Experts)
Llama 4 Maverick 17B FP8 (128 Experts)
Provider
Meta Llama
Meta Llama
Parameters
17B active / 16 experts
17B active / 128 experts (400B+ total)
Context window
10M tokens
1M tokens
Maximum output
Not listed
Not listed
Catalog pricing
Llama 4 Community License — free. Third-party: $0.10-0.30/MTok
Llama 4.1 Community License — free for most. Third-party APIs: $0.20-0.50/MTok
Listed input/output cost per 1M tokens
Separate numeric input/output rates not listed
Separate numeric input/output rates not listed
License
Llama 4 Community License
Llama 4.1 Community License
Coding support
Listed
Listed
Reasoning support
Not listed
Listed
Vision input
Listed
Listed
Multimodal
Listed
Listed
Open weights
Yes
Yes
Benchmark: MMLU
79.9%
85.5%
Benchmark: HumanEval
80.1%
88.0%
Benchmark: DocVQA
88.4%
91.6%
Figures are transcribed from LITHOVEX model records. Benchmark versions and evaluation setups may differ; pricing and specifications can change. Check each provider's current documentation before making a decision.
Frequently Asked Questions
What is different between Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts)?
Llama 4 Scout 17B (16 Experts): 10M tokens; Llama 4 Maverick 17B FP8 (128 Experts): 1M tokens. Llama 4 Scout 17B (16 Experts) has the larger listed context window. Llama 4 Scout 17B (16 Experts) is listed for coding, vision, multimodal input, agentic workflows, long-context tasks; Llama 4 Maverick 17B FP8 (128 Experts) is listed for coding, reasoning, vision, multimodal input, agentic workflows, long-context tasks. Shared catalog benchmark entries: MMLU: Llama 4 Scout 17B (16 Experts) 79.9%; Llama 4 Maverick 17B FP8 (128 Experts) 85.5%; HumanEval: Llama 4 Scout 17B (16 Experts) 80.1%; Llama 4 Maverick 17B FP8 (128 Experts) 88.0%; DocVQA: Llama 4 Scout 17B (16 Experts) 88.4%; Llama 4 Maverick 17B FP8 (128 Experts) 91.6%. These records are reference data, not a controlled head-to-head test; verify current pricing and specifications with each provider.
What benchmark values are listed for Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts)?
MMLU: Llama 4 Scout 17B (16 Experts) 79.9%; Llama 4 Maverick 17B FP8 (128 Experts) 85.5%; HumanEval: Llama 4 Scout 17B (16 Experts) 80.1%; Llama 4 Maverick 17B FP8 (128 Experts) 88.0%; DocVQA: Llama 4 Scout 17B (16 Experts) 88.4%; Llama 4 Maverick 17B FP8 (128 Experts) 91.6%. The catalog values may use different versions and evaluation setups, so they are reference points rather than a controlled head-to-head result.
What are the listed costs per million tokens for Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts)?
Llama 4 Scout 17B (16 Experts): Separate numeric input/output rates not listed (Llama 4 Community License — free. Third-party: $0.10-0.30/MTok). Llama 4 Maverick 17B FP8 (128 Experts): Separate numeric input/output rates not listed (Llama 4.1 Community License — free for most. Third-party APIs: $0.20-0.50/MTok). Where separate numeric input and output rates are available, the worked example uses 1M input plus 1M output tokens. Confirm current terms with each provider.
How far apart are the listed release dates for Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts)?
The catalog lists Llama 4 Scout 17B (16 Experts) as April 2025 and Llama 4 Maverick 17B FP8 (128 Experts) as April 2025, about 0 months apart at the precision supplied.
When was the comparison data last updated?
The model catalog records 2026-10-16 as its data update date. Provider specifications and prices may have changed since then.
How should I choose between Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts)?
Compare the listed specifications and capabilities, review the benchmark notes, then test both with the same representative prompt for your task. This page does not claim a universal winner.
Create your free LITHOVEX account and compare Llama 4 Scout 17B (16 Experts) and Llama 4 Maverick 17B FP8 (128 Experts) on your actual tasks. No credit card, no API key, no subscriptions needed.