LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Llama 4 Maverick 17B Instruct

meta·meta/llama-4-maverick-17b-instruct·GA·Open weights·llama series
Open multimodal Llama for strong reasoning with efficient everyday serving

Specs & pricing

Input / output per 1M tokens
Official price
—
Lowest paid·AbacusGateway
$0.14 / $0.59
Blended $0.25 · 5.1× spread
Context
1,000,000
Output limit
8,192
Knowledge cutoff
2024-08
Released / updated
2025-04-05 / 2025-04-05
Capabilities
Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImage
Weights
Hugging Face

Llama 4 Maverick 17B Instruct by meta is offered by 18 providers on this page. No official API price is listed for it; the prices below come from third-party channels. The lowest paid channel is Abacus at $0.14 / $0.59 per 1M. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it. Prices vary about 5.1× across the providers that host it.

Median output speed is 180 tokens per second, with 0.98s to the first token.

The context window is 1,000,000 tokens at the reference host, but hosts report different limits, from 8,192 to 1,048,576, so the usable window depends on the provider you pick. It supports tool use and structured output. Accepted input modalities are Text and Image. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2024-08.

Quality & performanceArtificial Analysis · Intelligence Index v5

Intelligence—
Value—
Output speed180 tok/s
TTFT0.98 s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 18 providers18 with public prices

  • ProviderInput / Output
  • AbacusGateway
    meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
    $0.14 / $0.59
    Context 1,048,576 ⚠Output limit 8,192deprecated
  • HeliconeGateway
    llama-4-maverick
    $0.15 / $0.60
    Context 131,072 ⚠Output limit 8,192
  • Deep InfraCloud
    meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
    $0.20 / $0.80
    Context 1,048,576 ⚠Output limit 16,384deprecated
  • DigitalOceanCloud
    llama-4-maverick
    $0.25 / $0.87
    Cache read $0.05Context 128,000 ⚠Output limit 16,384
  • IO.NETCloud
    meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
    $0.27 / $0.83
    Cache read $0.13Context 430,000 ⚠Output limit 4,096
  • LLM GatewayGateway
    $0.27 / $0.85
    Context 1,048,576 ⚠Output limit 8,192
  • DevPass (LLM Gateway)Gateway
    $0.27 / $0.85
    Context 1,048,576 ⚠Output limit 2,048
  • Novita AIGateway
    meta-llama/llama-4-maverick-17b-128e-instruct-fp8
    $0.27 / $0.85
    Context 1,048,576 ⚠Output limit 8,192
  • LLM GatewayGateway
    $0.24 / $0.97
    Context 8,192 ⚠Output limit 2,048deprecated
  • Amazon BedrockCloud
    meta.llama4-maverick-17b-instruct-v1:0
    $0.24 / $0.97
    Context 1,000,000Output limit 8,192deprecated
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Abacus
meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
Gateway$0.14$0.59——1,048,576 ⚠8,192deprecated
Helicone
llama-4-maverick
Gateway$0.15$0.60——131,072 ⚠8,192
Deep Infra
meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
Cloud$0.20$0.80——1,048,576 ⚠16,384deprecated
DigitalOcean
llama-4-maverick
Cloud$0.25$0.87$0.05—128,000 ⚠16,384
IO.NET
meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
Cloud$0.27$0.83$0.13—430,000 ⚠4,096
LLM GatewayGateway$0.27$0.85——1,048,576 ⚠8,192
DevPass (LLM Gateway)Gateway$0.27$0.85——1,048,576 ⚠2,048
Novita AI
meta-llama/llama-4-maverick-17b-128e-instruct-fp8
Gateway$0.27$0.85——1,048,576 ⚠8,192
LLM GatewayGateway$0.24$0.97——8,192 ⚠2,048deprecated
Amazon Bedrock
meta.llama4-maverick-17b-instruct-v1:0
Cloud$0.24$0.97——1,000,0008,192deprecated

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

Common scenarios
1Abacus$57.50
2Helicone$60.00
3DigitalOcean$69.50
4IO.NET$78.94
5Deep Infra$80.00
6LLM Gateway$96.50
The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.25$02026-08-052026-10-23
2026-10-23Input list —Output list —Min blended $0.25

Benchmark2 items

NameConditionsScoreMetricSource
Aider Polyglot—15.6percent correctSource ↗
SWE-Bench Prodataset: public5.24resolve rateSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Artificial Analysis evaluations18 items

GPQA Diamond67.1%
Humanity's Last Exam4.9%
MMMU-Pro62.1%
CritPt0.0%
SciCode31.7%
LiveCodeBench39.7%
Terminal-Bench Hard6.8%
Terminal-Bench 2.17.9%
Terminal-Bench 4.00.0%
AIME 202519.3%
τ²-Bench Telecom17.8%
τ³-Bench Banking3.7%
AutomationBench-AA0.2%
AA-Briefcase0
GDPval-AA-268
GDP.pdf0.8%
AA-LCR50.0%
IFBench43.0%

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and scale (percent correct, points or an index), so rows are not comparable with one another.

Related models

Omega Directive 24B Unslop v2.0same series—EVA-LLaMA-3.33-70B-v0.1same series—Llama 3.3 70B Instruct abliteratedsame series—Llama-3.3-70B-Instructsame lab—Llama 4 Scout 17B Instructsame lab—
Data partly from models.dev (MIT)