Arxiv:2504.07491 Base model:finetune:moonshotai... Base model:moonshotai/kimi-vl-... Conversational Custom code Feature-extraction Image-text-to-text Instruct Kimi vl Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/moonshotai/Kimi-VL-A3B-Thinking
Kimi VL A3B Thinking Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Kimi VL A3B Thinking Parameters and Internals
| LLM Name | Kimi VL A3B Thinking |
| Repository 🤗 | https://huggingface.co/moonshotai/Kimi-VL-A3B-Thinking |
| Base Model(s) | |
| Model Size | 16.4b |
| Required VRAM | 32.8 GB |
| Updated | 2025-04-23 |
| Maintainer | moonshotai |
| Model Type | kimi_vl |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | KimiVLForConditionalGeneration |
| License | mit |
| Model Max Length | 1048576 |
| Transformers Version | 4.50.3 |
| Tokenizer Class | TikTokenTokenizer |
| Padding Token | [PAD] |
| Vocabulary Size | 163840 |
| Torch Data Type | bfloat16 |
Rank the Kimi VL A3B Thinking Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 46599 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227