Llama 4 Maverick · Meta

What does Llama 4 Maverick cost to run in a coding agent?

Llama 4 Maverick costs about $0.0095 per average coding-agent request on our measured traffic, at OpenRouter's listed prices of $0.188 per million input and $0.653 per million output tokens. It scores 10 on the Artificial Analysis Intelligence Index (an estimate), 46th of the 48 models we track. Source: Artificial Analysis (artificialanalysis.ai)

List prices: OpenRouter listed price, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. How we compute this

Run current Claude, GPT, Gemini and open models in your agent

Or let the router send each step to the cheapest model that can do it, on providers we vet for price, uptime and speed. Plans from $19 a month.

See plans

What Llama 4 Maverick costs per coding-agent request

Cost at list price	Llama 4 Maverick
Prompt under 100K tokens50% of requests, 87% of input cached	$0.0043
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.012
Prompt over 272K tokens11% of requests, 86% of input cached	$0.026
Average request	$0.0095
100 requests	$0.95
Median session16 requests; a session ends after 10 idle minutes	$0.11
A month of regular usemedian of 20 users active on 5+ days	$15.76

Each logged request priced at the listed rates above for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request.

Llama 4 Maverick benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	10 (estimate)	46 of 48
Terminal-Bench 4.0agentic coding in a terminal	0.0%	35 of 39
Output speedmedian	44 tokens/s	40 of 44
Time to first answer tokenincludes thinking time	1.1 s	2 of 44

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: Llama 4 Maverick. Fetched October 13, 2026. Rank is among the models on this site, older versions included. Artificial Analysis runs Terminal-Bench 4.0 in its own setup, so its scores differ from the tbench.ai leaderboard.

About Llama 4 Maverick

Llama 4 Maverick is Meta's open-weight mixture-of-experts model from April 5, 2025, with a 1M-token context and support across most local runtimes and tooling. It is well behind current models: Artificial Analysis estimates 10 on its Intelligence Index and records 0% in its Terminal-Bench 4.0 run, at about 44 output tokens per second. At OpenRouter's listed price it costs $0.0095 per request in our measurements.

Sources: Artificial Analysis

Llama 4 Maverick FAQ

How much does Llama 4 Maverick cost per coding-agent request?

$0.0095 on average at OpenRouter's listed price, $0.95 per 100 requests, and $0.11 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at OpenRouter's listed rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does Llama 4 Maverick charge more for long prompts?

No. Llama 4 Maverick bills the same rate per token up to its 1M-token context window.

How fast is Llama 4 Maverick?

Artificial Analysis measures Llama 4 Maverick at 44 output tokens per second, with 1.1 seconds to the first answer token, thinking time included. Source: Artificial Analysis (artificialanalysis.ai).

Is Llama 4 Maverick good for AI agents?

Self-hosted setups built on the Llama ecosystem. On our scoring, which is calibrated to public benchmarks, it rates 4.4/10 for agentic ability and 3.5/10 for output quality. Open weights with wide runtime support.

What is Llama 4 Maverick's context window?

Llama 4 Maverick has a 1.0M-token context window, large enough to hold sizeable codebases or document sets in a single run. Meta is the maker.

Is Llama 4 Maverick open source?

Yes. Llama 4 Maverick ships with open weights, so you can self-host it or run it through a hosted provider such as OpenRouter. Check the license before commercial use.

What are the downsides of Llama 4 Maverick?

The main trade-offs: Not suited to agentic coding (0% in Artificial Analysis's Terminal-Bench 4.0 run); About 44 output tokens/s, per Artificial Analysis. It's strongest for self-hosted setups built on the Llama ecosystem.