Meta
Llama 3.2 3B
Open Llama instruction model for multilingual chat, reasoning, and coding
Model facts
- Context window
- 131K tokens
- Maximum output
- 32K tokens
- Cheapest paid input
- $0.02 per 1M tokens
- Cheapest paid output
- $0.02 per 1M tokens
- Open weights
- yes
- Providers
- 11
Capabilities and modalities
reasoning, tool calling, structured output, attachments, temperature control, open weights, text, pdf.
Local hardware estimate
At 4K context: Q4_K_M 2.9 GiB, Q8_0 4.4 GiB, F16 7.2 GiB working memory.
- Parameters
- 3.2 billion
- Architecture
- llama
- Model license
- llama3.2
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| default | 3.9 | not measured | 3.3 | not measured |
Observed price history
- 2026-08-24: $0.02 input / $0.02 output per 1M tokens
API providers
- Nvidia — model id meta/llama-3.2-3b-instruct; free input / free output per 1M tokens; provider documentation
- Inference — model id meta/llama-3.2-3b-instruct; $0.02 input / $0.02 output per 1M tokens; provider documentation
- NanoGPT — model id meta-llama/llama-3.2-3b-instruct; $0.031 input / $0.049 output per 1M tokens; provider documentation
- DevPass (LLM Gateway) — model id llama-3.2-3b-instruct; $0.03 input / $0.05 output per 1M tokens; provider documentation
- NovitaAI — model id meta-llama/llama-3.2-3b-instruct; $0.03 input / $0.05 output per 1M tokens; provider documentation
- Pioneer — model id meta-llama/Llama-3.2-3B; $0.1 input / $0.1 output per 1M tokens; provider documentation
- Kilo Gateway — model id meta-llama/llama-3.2-3b-instruct; $0.05 input / $0.33 output per 1M tokens; provider documentation
- OpenRouter — model id meta-llama/llama-3.2-3b-instruct; $0.05 input / $0.33 output per 1M tokens; provider documentation
- Cloudflare Workers AI — model id @cf/meta/llama-3.2-3b-instruct; $0.051 input / $0.335 output per 1M tokens; provider documentation
- Pioneer — model id meta-llama/Llama-3.2-3B-Instruct; $0.1 input / $0.335 output per 1M tokens; provider documentation
- Venice AI — model id llama-3.2-3b; $0.15 input / $0.6 output per 1M tokens; provider documentation