Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -30,7 +30,7 @@ accross various devices, can be found [here](https://aihub.qualcomm.com/models/l
|
|
| 30 |
- Model-1 (Prompt Processor): Llama-PromptProcessor-Quantized
|
| 31 |
- Max context length: 1024
|
| 32 |
- Prompt processor input: 1024 tokens
|
| 33 |
-
- Prompt processor output:
|
| 34 |
- Model-2 (Token Generator): Llama-TokenGenerator-KVCache-Quantized
|
| 35 |
- Token generator input: 1 input token + past KVCache
|
| 36 |
- Token generator output: 1 output token + KVCache for next iteration
|
|
|
|
| 30 |
- Model-1 (Prompt Processor): Llama-PromptProcessor-Quantized
|
| 31 |
- Max context length: 1024
|
| 32 |
- Prompt processor input: 1024 tokens
|
| 33 |
+
- Prompt processor output: 1024 output tokens + KVCache for token generator
|
| 34 |
- Model-2 (Token Generator): Llama-TokenGenerator-KVCache-Quantized
|
| 35 |
- Token generator input: 1 input token + past KVCache
|
| 36 |
- Token generator output: 1 output token + KVCache for next iteration
|