ai-models/Qwen3.6-35B-A3B
2
0
Fork 0
mirror of https://www.modelscope.cn/Qwen/Qwen3.6-35B-A3B.git synced 2026-07-16 06:42:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

Upload README.md to ModelScope hub

Browse Source
This commit is contained in:
VoyagerX
2026-04-23 15:47:15 +00:00
parent 9add5e2736
commit a005c49d47
1 changed files with 5 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
README.md
View File

@ -667,8 +667,7 @@ export OPENAI_API_KEY="EMPTY"
> We recommend using the following set of sampling parameters for generation
> - Thinking mode for general tasks: `temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0`
> - Thinking mode for precise coding tasks (e.g. WebDev): `temperature=0.6, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, repetition_penalty=1.0`
> - Instruct (or non-thinking) mode for general tasks: `temperature=0.7, top_p=0.8, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0`
> - Instruct (or non-thinking) mode for reasoning tasks: `temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0`
> - Instruct (or non-thinking) mode: `temperature=0.7, top_p=0.80, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0`
>
> Please note that the support for sampling parameters varies according to inference frameworks.
@ -728,7 +727,7 @@ messages = [
}
]
response = client.chat.completions.create(
chat_response = client.chat.completions.create(
model="Qwen/Qwen3.6-35B-A3B",
messages=messages,
max_tokens=81920,
@ -773,7 +772,7 @@ messages = [
#
# By default, `fps=2` and `do_sample_frames=True`.
# With `do_sample_frames=True`, you can customize the `fps` value to set your desired video sampling rate.
response = client.chat.completions.create(
chat_response = client.chat.completions.create(
model="Qwen/Qwen3.6-35B-A3B",
messages=messages,
max_tokens=81920,
@ -1011,10 +1010,8 @@ To achieve optimal performance, we recommend the following settings:
`temperature=1.0`, `top_p=0.95`, `top_k=20`, `min_p=0.0`, `presence_penalty=1.5`, `repetition_penalty=1.0`
- **Thinking mode for precise coding tasks (e.g., WebDev)**:
`temperature=0.6`, `top_p=0.95`, `top_k=20`, `min_p=0.0`, `presence_penalty=0.0`, `repetition_penalty=1.0`
- **Instruct (or non-thinking) mode for general tasks**:
`temperature=0.7`, `top_p=0.8`, `top_k=20`, `min_p=0.0`, `presence_penalty=1.5`, `repetition_penalty=1.0`
- **Instruct (or non-thinking) mode for reasoning tasks**:
`temperature=1.0`, `top_p=1.0`, `top_k=40`, `min_p=0.0`, `presence_penalty=2.0`, `repetition_penalty=1.0`
- **Instruct (or non-thinking) mode**:
`temperature=0.7`, `top_p=0.80`, `top_k=20`, `min_p=0.0`, `presence_penalty=1.5`, `repetition_penalty=1.0`
- For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
2. **Adequate Output Length**: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 81,920 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.