When was LLaMA 4 Maverick released?

LLaMA 4 Maverick was released by Meta on Saturday, Apr 5 2025.

Who made LLaMA 4 Maverick?

LLaMA 4 Maverick was built by Meta. Develops the open-weight Llama series of models. Committed to open-source AI research.

What benchmark scores did LLaMA 4 Maverick get?

LLaMA 4 Maverick reports 1 tracked benchmark score — BullshitBench v2: 29%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

Is LLaMA 4 Maverick open source?

Partly. LLaMA 4 Maverick is an open-weight model: the trained weights are free to download and run locally or on your own infrastructure, but the training data and code are not fully released and the license may restrict some commercial uses. It is not open source in the strict sense.

What came before and after LLaMA 4 Maverick?

Meta's previous tracked release was LLaMA 4 Scout on Apr 5 2025. It was followed by Muse Spark on Apr 8 2026.

LLaMA 4 MaverickOpen Weight

Released Apr 5 2025

LLaMA 4 Maverick is an AI model released by Meta on Saturday, Apr 5 2025, the same day as LLaMA 4 Scout. It is an open-weight model — the trained weights are available to download and run. Benchmark results (shown below) cover BullshitBench v2.

Available from


DigitalOcean	—	$0.1875	$0.6525	128K	—
Novita	—	$0.27	$0.85	1M	fp8
Parasail	—	$0.35	$1.00	512K	fp8
Google	us-east5	$0.35	$1.15	512K	—

USD per 1M tokens

Benchmarks

BullshitBench v2

29%

#61 of 97

Compare LLaMA 4 Maverick with

LLaMA 4 Maverick

Suggested comparisons

LLaMA 4 MaverickvsMuse Spark Open LLaMA 4 MaverickvsGPT-6.1 Sol LLaMA 4 MaverickvsClaude Haiku 5.5 LLaMA 4 MaverickvsGemini 4 Argon LLaMA 4 MaverickvsGrok 4.7 LLaMA 4 MaverickvsDeepSeek-V4.1-Flash-1017 LLaMA 4 MaverickvsMistral Large 4 LLaMA 4 MaverickvsKimi K3 LLaMA 4 MaverickvsGLM-5.3-Flash LLaMA 4 MaverickvsQwen3.8-Max-0902 LLaMA 4 MaverickvsNemotron 3.5 Super VL

About

Llama 4 Maverick, released April 5, 2025 alongside the smaller Llama 4 Scout, moved the Llama line to a natively multimodal mixture-of-experts design — roughly 400B total parameters with 17B active per token. It launched with claims of beating GPT-4o and Gemini 2.0 Flash at far lower serving cost.

The release proved controversial: an experimental chat-tuned variant's LMArena ranking drew accusations of benchmark gaming, and the promised frontier-scale Behemoth model never shipped. Llama 4 ended up the final release of the Llama era — Meta's next generation arrived in April 2026 under the new Muse Spark name.

Frequently asked questions

: LLaMA 4 Maverick was released by Meta on Saturday, Apr 5 2025.
: LLaMA 4 Maverick was built by Meta. Develops the open-weight Llama series of models. Committed to open-source AI research.
: LLaMA 4 Maverick reports 1 tracked benchmark score — BullshitBench v2: 29%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.
: Partly. LLaMA 4 Maverick is an open-weight model: the trained weights are free to download and run locally or on your own infrastructure, but the training data and code are not fully released and the license may restrict some commercial uses. It is not open source in the strict sense.
: Meta's previous tracked release was LLaMA 4 Scout on Apr 5 2025. It was followed by Muse Spark on Apr 8 2026.