🐍 | Python library for langchain using RunPod API endpoint as LLM.
Project description
Installation
# Install the latest release version
pip install runpod-llm
# or
# Install the latest development version (main branch)
pip install git+https://https://github.com/tsangwailam/langchain-runpod-llm
Get Runpod API key
- Goto www.runpod.io. Create a RunPod account.
- From the portal, goto Settings>APIKeys
- Create a new API key by click the "+ API Key" button.
Usage
from runpod_llm import RunpodLlama2
llm = RunpodLlama2(
apikey="YOU_RUNPOD_API_KEY",
llm_type="7b|13b",
config={
"max_tokens": 500,
#Maximum number of tokens to generate per output sequence.
"n": 1, # Number of output sequences to return for the given prompt.
"best_of": 1, # Number of output sequences that are generated from the prompt. From these best_of sequences, the top n sequences are returned. best_of must be greater than or equal to n. This is treated as the beam width when use_beam_search is True. By default, best_of is set to n.
"Presence penalty": 0.2, # Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.
"Frequency penalty": 0.5, # Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.
"temperature": 0.3, # Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling.
"top_p": 1, # Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens.
"top_k": -1, # Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens.
"use_beam_search": False, # Whether to use beam search instead of sampling.
},
verbose=True, # verbose output
)
some_prompt_template = xxxxx
output_chain = some_prompt_template | llm
output_chain.invoke({"input":"some input to prompt template"})
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for langchain-runpod-llm-0.0.2a0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | fc76a70ac863e3b76d39dd90f12c893fd97361888098fa2abdb52cbf3baad56d |
|
MD5 | 1f83e85e5b27ec492744ac8beeadab05 |
|
BLAKE2b-256 | 33b12cca821212e1684b6584c963aca188e5d3850dc011618dd53e5ce5d73828 |
Close
Hashes for langchain_runpod_llm-0.0.2a0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0cf68c01c903edd2fef8be7fa2cdc2615fce0606a1a3595a2617f452a26307aa |
|
MD5 | 411751ccc947a2ddb7d12e9c7d489077 |
|
BLAKE2b-256 | 449e61e1293bc307a32f093abb5773e2d8a4d958b1f789aed5d11871f6455e93 |