Inference
Llama 3.2 3B Instruct
textOpen weightactive
Context window
16,000
Maximum output
4,096
Catalog last checked
10/15/2026
Capabilities
tool_call
Deployment
API
Modalities
text
Current pricing
Effective rows only. When two sources disagree, both are shown and neither is picked for you.
input token
$0.02 / 1M tokens
Price unchanged since 9/25/2026, last confirmed 10/15/2026 per models.dev catalog · not yet cross-checked against a second source
output token
$0.02 / 1M tokens
Price unchanged since 9/25/2026, last confirmed 10/15/2026 per models.dev catalog · not yet cross-checked against a second source
Where to use Llama 3.2 3B Instruct
9 providers in the catalog list this model. Resellers and gateways often charge a different price from the model maker; check their terms, rate limits and data policies as well as price.
| Provider | Input / 1M tokens | Output / 1M tokens |
|---|---|---|
| MetaFirst-party | $0.05 | $0.34 |
| NVIDIA | $0.00 | $0.00 |
| Inference(this page) | $0.02 | $0.02 |
| DevPass (LLM Gateway) | $0.03 | $0.05 |
| NanoGPT | $0.03 | $0.05 |
| Kilo Gateway | $0.05 | $0.33 |
| Cloudflare Workers AI | $0.05 | $0.34 |
| Openrouter | $0.05 | $0.34 |
| Pioneer | $0.10 | $0.34 |