How to use Gemini with txtai #843

igorlima · 2024-12-26T13:30:17Z

I've been exploring the possibilities of using Google Gemini with txtai, but I haven't found any references to Gemini in the documentation yet.

Is there a way to embed text in txtai using Gemini? The documentation references other LLMs, but Gemini seems missing.

Here's a snippet of what I've attempted using the litellm method, though I haven't had any success so far:

import os
os.environ["GEMINI_API_KEY"] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

from txtai import LLM
llm = LLM("gemini/gemini-pro", method="litellm")
llm("Where is one place you'd go in Washington, DC?")

from txtai import Embeddings
embeddings = Embeddings(path="litellm/gemini/gemini-pro")
data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
]
embeddings.index(data)

I'd appreciate the guidance if anyone has insights or knows of any documentation or examples on using Gemini with txtai.

The text was updated successfully, but these errors were encountered:

davidmezzetti · 2024-12-26T15:38:28Z

Hello, first off thank you for the kind words!

What kind of error are you receiving?

igorlima · 2024-12-26T18:01:14Z

Before jumping into the error I'm facing, let me share my setup. I use a Docker container running Python 3.12.2 on MacOS.

here's a quick peek at my setup:
- I start the Docker container with this command:
```
docker run \
  --name python-learning --rm \
  --mount src=`realpath .`,target=/home/local,type=bind \
  --workdir /home/local \
  -it python:3.12.2 bash
```
- Since the Docker container is Linux-based, I needed to install libsndfile. You can find more info here and here. Here's a handy command to get it installed:
```
apt update
apt install libsndfile1
```

a Python script (main.py):

I've been playing around with the following Python script (main.py) to test the Gemini LLM model. Here's a snippet:

import os
os.environ["GEMINI_API_KEY"] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

from txtai import LLM
llm = LLM("gemini/gemini-pro", method="litellm")
llm("Where is one place you'd go in Washington, DC?")

from txtai import Embeddings
embeddings = Embeddings(path="litellm/gemini/gemini-pro")

data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
  "Beijing mobilises invasion craft along coast as Taiwan tensions escalate",
  "The National Park Service warns against sacrificing slower friends in a bear attack",
  "Maine man wins $1M from $25 lottery ticket",
  "Make huge profits without work, earn up to $100,000 a day"
]

embeddings.index(data)
print("%-20s %s" % ("Query", "Best Match"))
print("-" * 50)
for query in ("feel good story", "climate change", "public health story", "war", "wildlife", "asia", "lucky", "dishonest junk"):
  uid = embeddings.search(query, 1)[0][0]
  print("%-20s %s" % (query, data[uid]))

requirements.txt

# python3 -m venv my-env
# source my-env/bin/activate
#
# pip3 show txtai
# pip3 index versions txtai
# pip3 index versions txtai | grep -E "[(][0-9]+([.][0-9]+)+[)]"
# pip3 list
#
# pip3 install --no-cache --upgrade-strategy eager -I txtai==8.1.0
#
# pip3 install -r requirements.txt
#
txtai==8.1.0
model2vec==0.3.3
pudb==2024.1.3

# For macOS, there's a runtime dependency on `libomp`.
# Run `brew install libomp` in this case.
#
# For Linux, there's a runtime dependency on `libsndfile`.
# https://github.com/bastibe/python-soundfile#installation
# https://github.com/libsndfile/libsndfile/issues/297
# ```
# apt update
# apt search libsndfile
# apt install libsndfile1
# apt install libsndfile1-dev
# ```
txtai[pipeline]

After setting everything up and running python3 main.py, I ran into this error:

the entire log error

Give Feedback / Get Help: https://github.com/BerriAI/litellm/issues/new
LiteLLM.Info: If you need to debug this error, use `litellm.set_verbose=True'.

Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/litellm/main.py", line 2262, in completion
    response = vertex_chat_completion.completion(  # type: ignore
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py", line 1228, in completion
    data = sync_transform_request_body(**transform_request_params)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 379, in sync_transform_request_body
    return _transform_request_body(
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 342, in _transform_request_body
    raise e
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 307, in _transform_request_body
    content = litellm.GoogleAIStudioGeminiConfig()._transform_messages(
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/gemini/chat/transformation.py", line 131, in _transform_messages
    return _gemini_convert_messages_with_history(messages=messages)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 262, in _gemini_convert_messages_with_history
    raise e
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 253, in _gemini_convert_messages_with_history
    raise Exception(
Exception: Invalid Message passed in - {'content': "Where is one place you'd go in Washington, DC?", 'role': 'prompt'}. File an issue https://github.com/BerriAI/litellm/issues

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/local/python/2024a11m24d-txtai-embeddings/main-gemini.py", line 32, in <module>
    llm("Where is one place you'd go in Washington, DC?")
  File "/usr/local/lib/python3.12/site-packages/txtai/pipeline/llm/llm.py", line 63, in __call__
    return self.generator(text, maxlength, stream, stop, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/txtai/pipeline/llm/generation.py", line 54, in __call__
    results = self.execute(texts, maxlength, stream, stop, **kwargs)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/txtai/pipeline/llm/generation.py", line 86, in execute
    return list(self.stream(texts, maxlength, stream, stop, **kwargs))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/txtai/pipeline/llm/litellm.py", line 60, in stream
    result = api.completion(
             ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/utils.py", line 987, in wrapper
    raise e
  File "/usr/local/lib/python3.12/site-packages/litellm/utils.py", line 868, in wrapper
    result = original_function(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/main.py", line 3012, in completion
    raise exception_type(
          ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 2146, in exception_type
    raise e
  File "/usr/local/lib/python3.12/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 2122, in exception_type
    raise APIConnectionError(
litellm.exceptions.APIConnectionError: litellm.APIConnectionError: Invalid Message passed in - {'content': "Where is one place you'd go in Washington, DC?", 'role': 'prompt'}. File an issue https://github.com/BerriAI/litellm/issues
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/litellm/main.py", line 2262, in completion
    response = vertex_chat_completion.completion(  # type: ignore
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py", line 1228, in completion
    data = sync_transform_request_body(**transform_request_params)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 379, in sync_transform_request_body
    return _transform_request_body(
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 342, in _transform_request_body
    raise e
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 307, in _transform_request_body
    content = litellm.GoogleAIStudioGeminiConfig()._transform_messages(
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/gemini/chat/transformation.py", line 131, in _transform_messages
    return _gemini_convert_messages_with_history(messages=messages)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 262, in _gemini_convert_messages_with_history
    raise e
  File "/usr/local/lib/python3.12/site-packages/litellm/llms/vertex_ai/gemini/transformation.py", line 253, in _gemini_convert_messages_with_history
    raise Exception(
Exception: Invalid Message passed in - {'content': "Where is one place you'd go in Washington, DC?", 'role': 'prompt'}. File an issue https://github.com/BerriAI/litellm/issues

image A image B

In short, it outputs the following error message:

Exception: Invalid Message passed in - {'content': "Where is one place you'd go in Washington, DC?", 'role': 'prompt'}. File an issue https://github.com/BerriAI/litellm/issues

davidmezzetti · 2024-12-26T20:25:43Z

Looks like you're running into the same issue as this: #841

The original LLM pipeline was designed to work with raw prompts (i.e. with chat templates manually applied). Over time this has changed where chat messages are more the default. The issue above added a new defaultrole parameter to address this. At some point soon, the switch will be flipped to make the defaultrole="user" and issues like this will go away.

Nonetheless, here are the options in the meantime.

Pass chat messages instead of strings to the LLM pipeline.

llm([{"role": "user", "content": "Where is one place you'd go in Washington, DC?"}])

Install txtai from GitHub and set the defaultrole

llm("Where is one place you'd go in Washington, DC?", defaultrole="user")

Now with the Embeddings, all of the supported models are documented here: https://docs.litellm.ai/docs/embedding/supported_embedding

I'm not exactly sure if Gemini models generate embeddings and/or if they are supported by LiteLLM though.

igorlima · 2024-12-27T01:16:26Z

Absolutely; thank you so much for the helpful hint! 🙏 🕺

The resource shared in the previous comment was a game-changer. It successfully guided me through running the Gemini LLM pipeline and generating embeddings using Gemini. Here's a snippet of the code that worked seamlessly with the Gemini LLM model:

Gemini code snippet

import os
os.environ["GEMINI_API_KEY"] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

from txtai import LLM
# https://neuml.github.io/txtai/install/
# https://neuml.github.io/txtai/pipeline/text/llm/#example
# https://ai.google.dev/gemini-api/docs/models/gemini
llm = LLM("gemini/gemini-pro", method="litellm")
# print(llm("Where is one place you'd go in Washington, DC?", defaultrole="user"))
print(llm([{"role": "user", "content": "Where is one place you'd go in Washington, DC?"}]))

from txtai import Embeddings
# https://neuml.github.io/txtai/embeddings/configuration/vectors/#method
# https://docs.litellm.ai/docs/providers/gemini
# https://github.com/BerriAI/litellm/tree/12c4e7e695edb07d403dd14fc768a736638bd3d1/litellm/llms/vertex_ai
# https://github.com/BerriAI/litellm/blob/e19bb55e3b4c6a858b6e364302ebbf6633a51de5/model_prices_and_context_window.json#L2625
embeddings = Embeddings(path="gemini/text-embedding-004", method="litellm")

# works with a list, dataset or generator
data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
  "Beijing mobilises invasion craft along coast as Taiwan tensions escalate",
  "The National Park Service warns against sacrificing slower friends in a bear attack",
  "Maine man wins $1M from $25 lottery ticket",
  "Make huge profits without work, earn up to $100,000 a day"
]

# create an index for the list of text
embeddings.index(data)
print("%-20s %s" % ("Query", "Best Match"))
print("-" * 50)
# run an embeddings search for each query
for query in ("feel good story", "climate change", "public health story", "war", "wildlife", "asia", "lucky", "dishonest junk"):
  # extract uid of first result
  # search result format: (uid, score)
  uid = embeddings.search(query, 1)[0][0]
  # print text
  print("%-20s %s" % (query, data[uid]))

Additionally, I’ve prepared four more samples for other models: VertexAI, Mistral, Cohere, and AWS Bedrock. Anyone searching for examples of these models can find them right here.

other LLM models

VertexAI code snippet

import os
os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "application_default_credentials.json"

import litellm
# https://docs.litellm.ai/docs/embedding/supported_embedding#usage---embedding
# https://docs.litellm.ai/docs/providers/vertex
litellm.vertex_project = "hardy-device-38811" # Your Project ID
litellm.vertex_location = "us-central1"  # proj location

from txtai import LLM
# https://neuml.github.io/txtai/install/
# https://neuml.github.io/txtai/pipeline/text/llm/#example
llm = LLM("vertex_ai/gemini-pro", method="litellm")

# print(llm("Where is one place you'd go in Washington, DC?", defaultrole="user"))
print(llm([{"role": "user", "content": "Where is one place you'd go in Washington, DC?"}]))

from txtai import Embeddings
# https://neuml.github.io/txtai/embeddings/configuration/vectors/#method
# https://docs.litellm.ai/docs/providers/vertex
# embeddings = Embeddings(path="text-embedding-004", method="litellm")
embeddings = Embeddings(path="vertex_ai/text-embedding-004", method="litellm")

# works with a list, dataset or generator
data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
  "Beijing mobilises invasion craft along coast as Taiwan tensions escalate",
  "The National Park Service warns against sacrificing slower friends in a bear attack",
  "Maine man wins $1M from $25 lottery ticket",
  "Make huge profits without work, earn up to $100,000 a day"
]

# create an index for the list of text
embeddings.index(data)
print("%-20s %s" % ("Query", "Best Match"))
print("-" * 50)
# run an embeddings search for each query
for query in ("feel good story", "climate change", 
    "public health story", "war", "wildlife", "asia",
    "lucky", "dishonest junk"):
  # extract uid of first result
  # search result format: (uid, score)
  uid = embeddings.search(query, 1)[0][0]
  # print text
  print("%-20s %s" % (query, data[uid]))

Mistral code snippet

import os
os.environ["MISTRAL_API_KEY"] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

from txtai import LLM
# https://neuml.github.io/txtai/install/
# https://neuml.github.io/txtai/pipeline/text/llm/#example
llm = LLM("mistral/mistral-tiny", method="litellm")
# print(llm("Where is one place you'd go in Washington, DC?", defaultrole="user"))
print(llm([{"role": "user", "content": "Where is one place you'd go in Washington, DC?"}]))

import litellm
litellm.set_verbose=False

from txtai import Embeddings
# https://neuml.github.io/txtai/embeddings/configuration/vectors/#method
# https://docs.litellm.ai/docs/providers/mistral
embeddings = Embeddings(path="mistral/mistral-embed", method="litellm")

# works with a list, dataset or generator
data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
  "Beijing mobilises invasion craft along coast as Taiwan tensions escalate",
  "The National Park Service warns against sacrificing slower friends in a bear attack",
  "Maine man wins $1M from $25 lottery ticket",
  "Make huge profits without work, earn up to $100,000 a day"
]

# create an index for the list of text
embeddings.index(data)
print("%-20s %s" % ("Query", "Best Match"))
print("-" * 50)
# run an embeddings search for each query
for query in ("feel good story", "climate change", 
    "public health story", "war", "wildlife", "asia",
    "lucky", "dishonest junk"):
  # extract uid of first result
  # search result format: (uid, score)
  uid = embeddings.search(query, 1)[0][0]
  # print text
  print("%-20s %s" % (query, data[uid]))

Cohere code snippet

import os
os.environ["COHERE_API_KEY"] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

from txtai import LLM
# https://neuml.github.io/txtai/install/
# https://neuml.github.io/txtai/pipeline/text/llm/#example
llm = LLM("command-r", method="litellm")

# print(llm("Where is one place you'd go in Washington, DC?", defaultrole="user"))
print(llm([{"role": "user", "content": "Where is one place you'd go in Washington, DC?"}]))


from txtai import Embeddings
# https://neuml.github.io/txtai/embeddings/configuration/vectors/#method
# https://docs.litellm.ai/docs/providers/cohere
embeddings = Embeddings(path="cohere/embed-english-v3.0", method="litellm")

# works with a list, dataset or generator
data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
  "Beijing mobilises invasion craft along coast as Taiwan tensions escalate",
  "The National Park Service warns against sacrificing slower friends in a bear attack",
  "Maine man wins $1M from $25 lottery ticket",
  "Make huge profits without work, earn up to $100,000 a day"
]

# create an index for the list of text
embeddings.index(data)
print("%-20s %s" % ("Query", "Best Match"))
print("-" * 50)
# run an embeddings search for each query
for query in ("feel good story", "climate change", 
    "public health story", "war", "wildlife", "asia",
    "lucky", "dishonest junk"):
  # extract uid of first result
  # search result format: (uid, score)
  uid = embeddings.search(query, 1)[0][0]
  # print text
  print("%-20s %s" % (query, data[uid]))

AWS Bedrock code snippet

import os
os.environ["AWS_ACCESS_KEY_ID"] = "xxxxxxxxxxxxxxxxxxxx"  # Access key
os.environ["AWS_SECRET_ACCESS_KEY"] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" # Secret access key
os.environ["AWS_REGION_NAME"] = "us-west-2" # us-east-1, us-east-2, us-west-1, us-west-2

from txtai import LLM
# https://neuml.github.io/txtai/install/
# https://neuml.github.io/txtai/pipeline/text/llm/#example
llm = LLM("bedrock/amazon.titan-text-lite-v1", method="litellm")

# print(llm("Where is one place you'd go in Washington, DC?", defaultrole="user"))
print(llm([{"role": "user", "content": "Where is one place you'd go in Washington, DC?"}]))


from txtai import Embeddings
# https://neuml.github.io/txtai/embeddings/configuration/vectors/#method
# https://docs.litellm.ai/docs/providers/bedrock
embeddings = Embeddings(path="amazon.titan-embed-text-v1", method="litellm")

# works with a list, dataset or generator
data = [
  "US tops 5 million confirmed virus cases",
  "Canada's last fully intact ice shelf has suddenly collapsed, forming a Manhattan-sized iceberg",
  "Beijing mobilises invasion craft along coast as Taiwan tensions escalate",
  "The National Park Service warns against sacrificing slower friends in a bear attack",
  "Maine man wins $1M from $25 lottery ticket",
  "Make huge profits without work, earn up to $100,000 a day"
]

# create an index for the list of text
embeddings.index(data)
print("%-20s %s" % ("Query", "Best Match"))
print("-" * 50)
# run an embeddings search for each query
for query in ("feel good story", "climate change", 
    "public health story", "war", "wildlife", "asia",
    "lucky", "dishonest junk"):
  # extract uid of first result
  # search result format: (uid, score)
  uid = embeddings.search(query, 1)[0][0]
  # print text
  print("%-20s %s" % (query, data[uid]))

requirements.txt

# python3 -m venv my-env
# source my-env/bin/activate
#
# pip3 show txtai
# pip3 index versions txtai
# pip3 index versions txtai | grep -E "[(][0-9]+([.][0-9]+)+[)]"
# pip3 list
#
# pip3 install --no-cache --upgrade-strategy eager -I txtai==8.1.0
#
# pip3 install -r requirements.txt
#
txtai==8.1.0
model2vec==0.3.3
pudb==2024.1.3

# For macOS, there's a runtime dependency on `libomp`.
# Run `brew install libomp` in this case.
#
# For Linux, there's a runtime dependency on `libsndfile`.
# https://github.com/bastibe/python-soundfile#installation
# https://github.com/libsndfile/libsndfile/issues/297
# ```
# apt update
# apt search libsndfile
# apt install libsndfile1
# apt install libsndfile1-dev
# ```
txtai[pipeline]

# to use Vertex AI
# https://github.com/BerriAI/litellm/issues/5483
google-cloud-aiplatform==1.75.0

# to use AWS Bedrock
# https://docs.litellm.ai/docs/providers/bedrock
boto3==1.35.88

Once again, thank you for your support and guidance! 🙌 🎉

davidmezzetti · 2024-12-27T02:23:38Z

I'm glad this worked!

One minor thing, you shouldn't need method='litellm' but it doesn't hurt either.

I appreciate you documenting how to do this with all the model endpoints you did, thank you!

igorlima · 2024-12-27T04:10:04Z

This issue has not only helped improve the documentation for the litellm library but also brought an opportunity to enhance the documentation here.

Below is a proposed PR to add these embedding usages in the documentation.

Enhanced Docs: LLM Embedding Examples #844

This was referenced Dec 27, 2024

Update Documentation - Gemini Embedding BerriAI/litellm#7436

Merged

Enhanced Docs: LLM Embedding Examples #844

Merged

This was referenced Dec 30, 2024

Supporting Vision models from Groq getomni-ai/zerox#65

Open

supporting Gemini getomni-ai/zerox#74

Open

davidmezzetti assigned igorlima Jan 2, 2025

davidmezzetti added this to the v8.2.0 milestone Jan 2, 2025

davidmezzetti closed this as completed Jan 2, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

How to use Gemini with txtai #843

How to use Gemini with txtai #843

igorlima commented Dec 26, 2024 •

edited

Loading

davidmezzetti commented Dec 26, 2024

igorlima commented Dec 26, 2024 •

edited

Loading

davidmezzetti commented Dec 26, 2024

igorlima commented Dec 27, 2024

davidmezzetti commented Dec 27, 2024

igorlima commented Dec 27, 2024

How to use Gemini with txtai #843

How to use Gemini with txtai #843

Comments

igorlima commented Dec 26, 2024 • edited Loading

davidmezzetti commented Dec 26, 2024

igorlima commented Dec 26, 2024 • edited Loading

davidmezzetti commented Dec 26, 2024

igorlima commented Dec 27, 2024

davidmezzetti commented Dec 27, 2024

igorlima commented Dec 27, 2024

igorlima commented Dec 26, 2024 •

edited

Loading

igorlima commented Dec 26, 2024 •

edited

Loading