Model

Llama 4 Maverick 17B Instruct

Name: Llama 4 Maverick 17B Instruct
Brand: meta

Open multimodal Llama for strong reasoning with efficient everyday serving

6 providers · 9 listingsreleased Apr 5, 2025knowledge cutoff 2024-08open weights

$0.14

lowest input /M · Abacus

$0.59

lowest output /M · Abacus

1.05M

context window

43K

max output

Compare providers (sorted by listed input price)

6 providers list Llama 4 Maverick 17B Instruct. The cheapest full-context option is LLM Gateway at $0.27/$0.85 per M. Abacus lists lower but caps context at 1.05M.

			Cache read /M	Cache write /M		Capabilities
Abacus meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8	$0.14	$0.59	—	—	1.05M	no reasoningtoolsno structuredvision	Apr 5, 2025
LLM Gatewayvariant aws-bedrock/llama-4-maverick-17b-instruct	$0.24	$0.97	—	—	8K	no reasoningno toolsno structuredvision	Apr 5, 2025
Amazon Bedrockvariant us.meta.llama4-maverick-17b-instruct-v1:0	$0.24	$0.97	—	—	1M	no reasoningtoolsno structuredvision	Apr 5, 2025
Amazon Bedrockvariant meta.llama4-maverick-17b-instruct-v1:0	$0.24	$0.97	—	—	1M	no reasoningtoolsno structuredvision	Apr 5, 2025
Charm Hyper llama-4-maverick-17b-128e-instruct-fp8	$0.255	$0.837	$0.128	—	430K	no reasoningtoolsno structuredno vision	Jul 22, 2026
LLM Gatewayvariant novita/llama-4-maverick-17b-instruct	$0.27	$0.85	—	—	1.05M	no reasoningno toolsno structuredvision	Apr 5, 2025
DevPass (LLM Gateway) llama-4-maverick-17b-instruct	$0.27	$0.85	—	—	1.05M	no reasoningno toolsno structuredvision	Apr 5, 2025
Neon llama-4-maverick	$0.5	$1.50	—	—	1M	no reasoningtoolsstructuredvision	Apr 5, 2025
LLM Gatewayvariant scx-ai/llama-4-maverick-17b-instruct	$0.53	$1.62	—	—	131K	no reasoningtoolsno structuredno vision	Apr 5, 2025

Best price history

Daily lowest published price across live providers. A lower line is better for the buyer.

Best input /M: $0.14unchanged
Best output /M: $0.59unchanged
Provider coverage: 9; 9 at the start · 24 snapshots

Published best prices are unchanged.

24 daily snapshots collected from Aug 21, 2026 through Sep 14, 2026. A price chart will appear after the first real move.

Price badge

![Llama 4 Maverick 17B Instruct price](https://www.vaanalytics.in/badge/meta/llama-4-maverick-17b-instruct.svg)

https://www.vaanalytics.in/badge/meta/llama-4-maverick-17b-instruct.svg

Live SVG, regenerated on every hourly sync — paste it into a README to always show the current cheapest listed price. Updates when prices change; no build step needed on your side.

Recent changes

Changelog →

contextvia amazon-bedrock · Sep 13, 2026
- description: Open multimodal Llama model for strong reasoning and fast responses → Open multimodal Llama for strong reasoning with efficient everyday serving
- output: 16K → 8K tokens
repricedvia hyper · Sep 13, 2026
- input: $0.274 → $0.255
- output: $0.899 → $0.837
- cache_read: $0.137 → $0.128
contextvia amazon-bedrock · Sep 13, 2026
- output: 16K → 8K tokens
repricedvia hyper · Sep 11, 2026
- cache_read: — → $0.137
- cache_write: $0.137 → —
repricedvia hyper · Aug 31, 2026
- input: $0.284 → $0.274
- output: $0.934 → $0.899
- cache_write: $0.142 → $0.137
repricedvia hyper · Aug 28, 2026
- input: $0.274 → $0.284
- output: $0.899 → $0.934
- cache_write: $0.137 → $0.142

Benchmarks

Aider Polyglot ↗(percent correct)report ↗Independent15.6
SWE-Bench Pro ↗(resolve rate)report ↗Independent5.24

Scores a lab published about its own model are claims, not measurements. Rankings elsewhere on this site use independently measured scores only.

Weights

Hugging Face

Prices are per million tokens (USD). “—” means the provider does not publicly list a price for this model. Data from models.dev; verify with the provider before purchasing.