LlamaGuard 4 12B
LlamaGuard 4 12B is Meta's language model with a 164K context window and up to 16K output tokens, available from 4 providers, starting at $0.18 / 1M input and $0.18 / 1M output. A natively multimodal 12B safety classifier pruned from Llama 4 Scout, trained jointly on text and multiple images for comprehensive content moderation.
Capabilities
Input2/5
Text✓
Image✓
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities1/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs✓
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | |
|---|---|---|
| Input $ / 1M | Output $ / 1M | |
| $0.18 | $0.18 | |
| $0.2 | $0.2 | |
| $0.2 | $0.2 | |
| $0.18 | $0.18 | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host LlamaGuard 4 12B, ranked by cheapest on-demand price. The model needs about 29 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.
All clouds
FP16 (full precision)
US Dollar ($)
Instance | Cloud | GPU | VRAM | Price | Cheapest region | |
|---|---|---|---|---|---|---|
| Standard_NV16as_v4 |  Azure | AMD Radeon Instinct MI25 | 32 GB | $0.932/hr | westus2 | |
| Standard_NP10s | Azure | AMD Alveo U250 FPGA (64GB) | 64 GB | $1.65/hr | westus2 | |
| g6e.xlarge |  AWS | L40S | 45 GB | $1.86/hr | us-east-1 | |
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| LlamaGuard 4 12B | 164K | $0.180 | $0.180 | Current | |
| Llama 4 Maverick | 1.0M | $0.120 | $0.485 | Available | |
| Llama 4 Scout | 10.0M | $0.100 | $0.300 | Available |