Total runs:
42.5K
Run Growth:
22.8K
Growth Rate:
53.77%
Sponsored by
YouArt
Kwaipilot / KAT-Dev-FP8
Model's Last Updated: October 10 2025
text-generation
Introduction of KAT-Dev-FP8
Model Details of KAT-Dev-FP8
This repository contains an FP8 quantized version of the Kwaipilot/KAT-Dev model.
News
🔥 We’re thrilled to announce the release of KAT-Dev-72B-Exp, our latest and most powerful model yet!
🔥 You can now try our strongest proprietary coder model KAT-Coder directly on the StreamLake platform for free .
Highlights
KAT-Dev-32B is an open-source 32B-parameter model for software engineering tasks.
On SWE-Bench Verified, KAT-Dev-32B achieves comparable performance with 62.4% resolved and ranks 5th among all open-source models with different scales.
Introduction
KAT-Dev-32B is optimized via several stages of training, including a mid-training stage, supervised fine-tuning (SFT) & reinforcement fine-tuning (RFT) stage and an large-scale agentic reinforcement learning (RL) stage. In summary, our contributions include:
| Stage | Key Techniques |
|---|---|
| 1. Mid-Training | We observe that adding extensive training for tool-use capability, multi-turn interaction, and instruction-following at this stage may not yield large performance gains in the current results (e.g., on leaderboards like SWE-bench). However, since our experiments are based on the Qwen3-32B model, we find that enhancing these foundational capabilities will have a significant impact on the subsequent SFT and RL stages. This suggests that improving such core abilities can profoundly influence the model’s capacity to handle more complex tasks. |
| 2. SFT & RFT | We meticulously curated eight task types and eight programming scenarios during the SFT stage to ensure the model’s generalization and comprehensive capabilities. Moreover, before RL, we innovatively introduced an RFT stage. Compared with traditional RL, we incorporate “teacher trajectories” annotated by human engineers as guidance during training—much like a learner driver being assisted by an experienced co-driver before officially driving after getting a license. This step not only boosts model performance but also further stabilizes the subsequent RL training. |
| 3. Agentic RL Scaling | Scaling agentic RL hinges on three challenges: efficient learning over nonlinear trajectory histories, leveraging intrinsic model signals, and building scalable high-throughput infrastructure. We address these with a multi-level prefix caching mechanism in the RL training engine, an entropy-based trajectory pruning technique, and an inner implementation of SeamlessFlow[1] architecture that cleanly decouples agents from training while exploiting heterogeneous compute. These innovations together cut scaling costs and enable efficient large-scale RL. |
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog .
Quickstart
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Kwaipilot/KAT-Dev-FP8"
# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
# conduct text completion
generated_ids = model.generate(
**model_inputs,
max_new_tokens=65536
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
content = tokenizer.decode(output_ids, skip_special_tokens=True)
print("content:", content)
Claude Code
vllm server
MODEL_PATH="Kwaipilot/KAT-Dev-FP8"
vllm serve $MODEL_PATH \
--enable-prefix-caching \
--tensor-parallel-size 8 \
--tool-parser-plugin $MODEL_PATH/qwen3coder_tool_parser.py \
--chat-template $MODEL_PATH/chat_template.jinja \
--enable-auto-tool-choice --tool-call-parser qwen3_coder
claude-code-router
is a third-party routing utility that allows Claude Code to flexibly switch between different backend APIs.
On the dashScope platform, you can install the
claude-code-config
extension package, which automatically generates a default configuration for
claude-code-router
with built-in dashScope support.
Once the configuration files and plugin directory are generated, the environment required by
ccr
will be ready.
If needed, you can still manually edit
~/.claude-code-router/config.json
and the files under
~/.claude-code-router/plugins/
to customize the setup.
Finally, simply start
ccr
to run Claude Code and seamlessly connect it with the powerful coding capabilities of
KAT-Dev-32B
.
Happy coding!
Runs of Kwaipilot KAT-Dev-FP8 on huggingface.co
6
Total runs
0
24-hour runs
0
3-day runs
2
7-day runs
0
30-day runs
More Information About KAT-Dev-FP8 huggingface.co Model
Url of KAT-Dev-FP8
KAT-Dev-FP8 huggingface.co Url
Provider of KAT-Dev-FP8 huggingface.co
Other API from Kwaipilot
Updated:July 28 2026
Total runs:
871
Run Growth:
-802
Growth Rate:
-92.08%
Total runs:
268
Run Growth:
94
Growth Rate:
35.07%
Total runs:
84
Run Growth:
-57
Growth Rate:
-67.86%
Total runs:
23
Run Growth:
0
Growth Rate:
0.00%
Total runs:
15
Run Growth:
9
Growth Rate:
60.00%
Total runs:
14
Run Growth:
-20
Growth Rate:
-142.86%