1974376d70
* Add GLM4 MoE DSA model implementation with configurable parameters * Update Acknowledgments to include GLM4 MoE DSA support * format * update ackn. * Fixes * Update acknowledgments to include contributions for GLM MoE DSA and additional architectures * use dsv32 for glm5 * fix * Fix rope theta --------- Co-authored-by: Tarjei Mandt <[email protected]> Co-authored-by: Awni Hannun <[email protected]>
33 lines
2.1 KiB
Markdown
33 lines
2.1 KiB
Markdown
# Individual Contributors
|
|
|
|
If you wish to be acknowledged for your contributions, please list your name
|
|
with a short description of your contribution(s) below. For example:
|
|
|
|
- Jane Smith: Added the `foo` example.
|
|
|
|
MLX LM was developed with contributions from the following individuals:
|
|
|
|
- Shunta Saito: Added support for PLaMo models.
|
|
- Gökdeniz Gülmez: Added support for the following architectures:
|
|
OpenBMB's `MiniCPM` and `MiniCPM3`, Kyutai's `Helium`, State-Space's `Mamba v1` and
|
|
`Mamba v2`, Z.ai & THUKEG's `GLM`, `GLM4`, `GLM5 (GLM MoE DSA)`, Rednote `dots.llm1`, Baidu's `Ernie4.5 MoE`,
|
|
inclusionAI's `Bailing MoE e.g. Ling-family`, `Bailing MoE Linear e.g. Ling-Linear-family`,
|
|
Klear team - Kuaishou Technology's `Klear`, AI21 Lab's `Jamba` IBM's `Granite MoE`,
|
|
Meituan's `LongCat`, Nvidia's `Nemotron H`, Swiss-AI's `Apertus`, Nikity's `Lille130m`,
|
|
Alibaba Qwen's `Qwen3Next`, Tele-AI's `TeleChat3`, and Allenai's `OLMoE` and `Olmo 3`;
|
|
Helped add support for the following model architectures:
|
|
Alibaba Qwen's `Qwen3 & Qwen3MoE)`; Added support for the following training algorithms:
|
|
`Full Weight Fine-Tuning`, and the `Muon` optimizer;
|
|
Added support for the following other features:
|
|
`Multiple Optimizers to choose for training`, and `reporting training metrics to WandB (Weights & Biases)`.
|
|
- Prince Canuma: Helped add support for the following model architectures:
|
|
HuggingFace's `Starcoder2`, Cohere's `Cohere (1 and 2)`, Alibaba Qwen's `Qwen (2, 3 and MoE)`,
|
|
Microsoft's `Phi (3 and 3.5 MoE)`, `BitNet1.58`, Meta's `Llama (3 and 4)`, MinimaxAI's `MiniMax`,
|
|
MoonshotAI's `Kimi-Linear`, LiquidAI's `LFM2` and `LFM2 MoE`,
|
|
Google DeepMind's `Gemma 3`, TII's `Falcon H1` and InterLM's `InternLM 2.5`.
|
|
- Ivan Fioravanti: Added support for the following architectures:
|
|
ServiceNow-AI's `Apriel 1.5`, Tencent's `Hunyuan Dense V1` and `Hunyuan MoE V1`.
|
|
- Tarjei Mandt: Added support for the following architectures: `Step 3.5 Flash`,
|
|
MoonshotAI's `Kimi K2.5`, Upstage's `Solar Open`, LG AI Research's `K-Exaone MoE`,
|
|
Meituan's `LongCat Flash Lite` Helped add support for the following model architectures:
|
|
Z.ai & THUKEG's `GLM5 (GLM MoE DSA)` |