Llama 4 Behemoth
Meta's largest announced Llama, a roughly 2T-parameter MoE with 288B active across 16 experts, announced April 2025 and never released.
Key Specifications
- Inputs
- Text, image
- Capabilities Show details
- Function calling Connect to external tools, APIs, and systems.
Hosted API pricing
| Provider | Input /1M tokens | Output /1M tokens | Cost at 10M in + 2M out | |
|---|---|---|---|---|
|
Meta
Creator
|
Open weights, no hosted price | View | ||
Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.
Frequently Asked Questions
What is Llama 4 Behemoth good for?
Reference only: Meta announced Behemoth in April 2025 as a roughly 2T-parameter mixture of experts with 288B active across 16 experts, and never released it.
When is Llama 4 Behemoth not a good fit?
Any use. It was never released: no weights to download, no hosted API, and no published limits.
Can I self-host Llama 4 Behemoth?
No. Meta does not publish the weights; Llama 4 Behemoth is available only through hosted APIs.
More from Meta
| Model | Context | Input /1M | Output /1M |
|---|---|---|---|
|
Muse Glimmer 30B
Open weights
|
131K | $0.20 | $0.80 |
|
Llama 4 Scout
Open weights
|
10M | $0.16 | $0.64 |
|
Llama 4 Maverick
Open weights
|
1M | $0.20 | $0.70 |
|
Llama 3.1 8B
Open weights
|
131K | $0.02 | $0.05 |
|
Llama 3.2 11B Vision
Open weights
|
128K | $0.05 | $0.68 |
|
Llama 3.2 1B
Open weights
|
128K | $0.03 | $0.20 |
|
Llama 3.2 3B
Open weights
|
128K | $0.05 | $0.34 |
|
Llama 3.2 90B Vision
Open weights
|
128K | No hosted price | |