Model Card on HF 🤗: https://huggingface.co/Columbidae/Qwen3-16B-A3B-Base
Qwen3 16B A3B Base 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)
Qwen3 16B A3B Base Parameters and Internals
| LLM Name | Qwen3 16B A3B Base |
| Repository 🤗 | https://huggingface.co/Columbidae/Qwen3-16B-A3B-Base |
| Base Model(s) | |
| Model Size | 30b |
| Required VRAM | 32.1 GB |
| Updated | 2025-06-01 |
| Maintainer | Columbidae |
| Model Type | qwen3_moe |
| Model Files | |
| Model Architecture | Qwen3MoeForCausalLM |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.51.3 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives to Qwen3 16B A3B Base
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...0B A6B 16 Extreme 128K Context | 128K / 61.1 GB | 89 | 7 |
| Qwen3 30B A3B | 40K / 61.1 GB | 299905 | 612 |
| Qwen3 30B A3B FP8 | 40K / 32.5 GB | 196539 | 62 |
| Qwen3 16B A3B | 40K / 32.1 GB | 1765 | 80 |
| Qwen3 30B A6B 16 Extreme | 40K / 61.1 GB | 1309 | 49 |
| Qwen3 30B A3B Abliterated | 40K / 61.1 GB | 1580 | 31 |
| Qwen3 30B A3B.w4a16 | 40K / 16.7 GB | 1763 | 7 |
| ...en3 30B A3B Abliterated Erotic | 40K / 61.1 GB | 349 | 11 |
| Qwen3 30B A3B Bf16 | 40K / 61.2 GB | 743 | 2 |
| Qwen3 30B A1.5B High Speed | 40K / 61.1 GB | 315 | 8 |
Note: green Score (e.g. "73.2") means that the model is better than Columbidae/Qwen3-16B-A3B-Base .
Rank the Qwen3 16B A3B Base 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? 47770 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227