Llama 3.1 8B — Specifications
| Developer | Meta |
|---|
| Type | LLM (dense) |
|---|
| Modality | Text → Text |
|---|
| Parameters | 8B |
|---|
| Context window | 128K |
|---|
| Max output | — |
|---|
| License | Llama 3.1 Community (open) |
|---|
| Open weights | ✅ Yes |
|---|
| Released | 2024 |
|---|
| Input price | $0.02 /1M |
|---|
| Output price | $0.03 /1M |
|---|
| API providers | Together, DeepInfra, OpenRouter, Ollama |
|---|
🖥️ Run it locally
| VRAM (4-bit) | ~5 GB |
|---|
| Minimum GPU | Any 8GB GPU |
|---|
Official page →
The workhorse small Llama — 8B, 128K context, runs on almost any modern GPU. One of the most widely deployed open models in production, and an excellent cheap baseline.
