2.2k likesNew · 12d old
Open-source Qwen 3.8 Flash model now ready for vLLM serving
What is Qwen3.8-Flash-Next-E384-FP4-V2?
The model is a fully merged, 4-bit quantized version of Qwen-3.8-Flash that can turn images into text and answer prompts. It includes all weight shards, tokenizer, and a custom vLLM patch so you can launch it on a single B200 GPU with FP8 KV cache and tool-calling support.
Who it is for
If you need a low-memory, high-throughput multimodal LLM, you can drop this into your vLLM stack and start experimenting without extra LoRA loading.