Skip to main content

That's right. I'm Kira ✍️

Kira is a suite of LLMs built with JAX and Equinox.

It is designed to be as clean and simple as possible, while still being flexible and powerful. It also provides a simple training loop, which interoperates with the Jaxonloader library.


Available Models

Currently, Kira provides the following models: - Kira: A standard transformer, which allows for interpolation between MHA and MQA. - Mamba: The new selective state space model

Kira can also be used as an encoder. Simply pass mask=None when you call Kira and there will be no masking in the MHA (i.e. making it an encoder).


To get started with Kira, you can either install it with

pip3 install kira_llm

or simply clone the repository and cherry-pick what you need.

Metadata

Release files for kira_llm 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for kira_llm 0.2.0
File Size Uploaded
kira_llm-0.2.0.tar.gz 14.0 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for kira_llm 0.2.0
File Interpreter ABI Platform
kira_llm-0.2.0-py3-none-any.whl Python 3 none any Details

Total release size: 29.6 kB

Release files / kira_llm-0.2.0.tar.gz

Download URL kira_llm-0.2.0.tar.gz
Size 14.0 kB
Tags Source
SHA-256 checksum
How to use checksums
74fe8a39c866d061e46aa99aca01d18c6e241bbf9b4337225ae74ba58817e444
BLAKE2b-256 checksum
How to use checksums
113342d50967efcb35c1003f97a6c822cf6e9640520633a761fd8eaeef320cb5
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via python-httpx/0.26.0

Release files / kira_llm-0.2.0-py3-none-any.whl

Download URL kira_llm-0.2.0-py3-none-any.whl
Size 15.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
98cc068127831b5911799f7b179aa2e5c820fc6e59d24df306c71d17682e6fa9
BLAKE2b-256 checksum
How to use checksums
dc7197f450a00d95e9b79ff80e7df424b5e5183a6735c9bd7b12556423222e8d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via python-httpx/0.26.0

Release history Release notifications | RSS feed

This release

0.2.0 This release

2 release files

0.1.3

2 release files

0.1.2

2 release files

0.1.1

2 release files

0.1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page