- input
- $0.71
- cached
- $0.71
- output
- $0.71
- Context
- 131Ktokens
- Released
- 1y ago
- Size
- 70Bparameters
Cost/1M tokens
Open-weight builds
Ollama variants
Every published Ollama tag matched to this model, including parameter editions, instruction/base variants, and quantized builds.
| Variant | Parameters | Type | Quantization | Size | Context | Input |
|---|---|---|---|---|---|---|
| llama3.3:70b-instruct-fp16 | 70B | instruct | FP16 | 141 GB | 128K | text |
| llama3.3:70b-instruct-q2_K | 70B | instruct | Q2_K | 26 GB | 128K | text |
| llama3.3:70b-instruct-q3_K_M | 70B | instruct | Q3_K_M | 34 GB | 128K | text |
| llama3.3:70b-instruct-q3_K_S | 70B | instruct | Q3_K_S | 31 GB | 128K | text |
| llama3.3:70b-instruct-q4_0 | 70B | instruct | Q4_0 | 40 GB | 128K | text |
| llama3.3:latest | 70B | instruct | Q4_K_M | 43 GB | 128K | text |
| llama3.3:70b-instruct-q4_K_S | 70B | instruct | Q4_K_S | 40 GB | 128K | text |
| llama3.3:70b-instruct-q5_0 | 70B | instruct | Q5_0 | 49 GB | 128K | text |
| llama3.3:70b-instruct-q5_1 | 70B | instruct | Q5_1 | 53 GB | 128K | text |
| llama3.3:70b-instruct-q5_K_M | 70B | instruct | Q5_K_M | 50 GB | 128K | text |
| llama3.3:70b-instruct-q6_K | 70B | instruct | Q6_K | 58 GB | 128K | text |
| llama3.3:70b-instruct-q8_0 | 70B | instruct | Q8_0 | 75 GB | 128K | text |
Deployment graph
Can I run this?
Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.