RWKV7-G1b-2.9B-GGUF

1️⃣ What are G0 / G1 / G1a2 / G1b?

The fields like G0a / G1a / G1a2 in RWKV model names indicate versions of the training data. In terms of data quality, the ranking is: G1b > G1a3 > G1a2 > G1a > G1 > G0a2 > G0.

The RWKV7-G1a model is an advanced version of RWKV7-G1 that was further trained with 1T (1 trillion tokens) of high-quality inference and instruction data. RWKV7-G1a2 was produced by continuing to add more data and training on top of RWKV7-G1a. And so on.

[!TIP] More high-quality data will be added later to form the G1b dataset, and RWKV7-G1b series models will also be trained and open-sourced.

2️⃣ How to Choose the Best Model?

Look at the Date in the Model Name—with the same parameters, the newer the model, the better!

For instance, for the 1.5B models, the G1a2 version released on 251005 is definitely superior to the G1 version released on 250429.

3️⃣ What is the difference between the RWKV7-G series and the World series?

The RWKV7-G series supports an inference mode, which can be activated using the following format:

User: USER_PROMPT

Assistant: <think

How to choose the best model?

Look at the date in the model name — for the same parameter size, a newer model is better!

For example, for the same 1.5B model, a G1a2 version released on 251005 will definitely be superior to a G1 version released on 250429.

[!WARNING] For the 0.1B and 0.4B models, we recommend using FP16/Q8_0 quantization. Otherwise, the models may fail to complete tasks due to precision loss caused by quantization.

---

G0/G1/G1a2/G1b 是什么?

RWKV 模型名称中的 G0a/G1a/G1a2 等字段是训练数据的版本，数据质量排序：G1b > G1a3 > G1a2 > G1a > G1 > G0a2 > G0 。

RWKV7-G1a 模型是在 RWKV7-G1 模型的基础上继续训练了 1T 优质推理和指令数据的进阶版，RWKV7-G1a2 则是在 RWKV7-G1a 模型的基础上继续添加数据训练，以此类推。

[!TIP] 后续会继续添加优质数据形成 G1b 数据集，也会训练并开源 RWKV7-G1b 系列模型。

RWKV7-G 系列和 World 系列有什么区别？

RWKV7-G 系列模型支持推理模式，可通过以下格式开启推理模式：

User: USER_PROMPT

Assistant: <think

如何选择最好的模型？

看模型名称中的日期，相同的参数，模型越新越好！

比如同样是 1.5B 模型，发布于 251005 的 G1a2 版本必定优于 250429 的 G1 版本 。

[!WARNING] 对于 0.1B 和 0.4B 模型，我们建议使用 FP16/Q8_0 量化类型。否则模型可能因量化带来的精度损失而无法完成任务。