Qwen3.8-Flash-Next-E384-FP4-V2
Will it run on your hardware?
Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest
Need an exact figure for your context length? Use the VRAM calculator.
Run it locally
Copy-paste - running in under a minute
vllm serve autotrust/Qwen3.8-Flash-Next-E384-FP4-V2New to this? Start with Ollama · serve to many users with vLLM.
Deep dive
Notes, sources, and the full write-up
Qwen3.8-Flash-Next-E384-FP4-V2 is an open-weight vision-language model published by autotrust on Hugging Face in October 2026, with 155.1B parameters under a custom license (read the model card before commercial use). It is built on autotrust/Qwen3.8-Flash-Next-E384-FP4-Local.
Among the 74 vision-language models in this directory it ranks 46th by monthly Hugging Face downloads, and 20 of the others are larger.
HardwareLink to this section
Unquantized FP16 weights take about 310 GB; 8-bit brings that to ~166 GB and 4-bit (Q4_K_M) to ~90 GB, before activations and any context or image buffers. Even at 4-bit it needs more than 12 GB, so plan on a multi-GPU or high-memory Mac setup. Check exact numbers for your card with the VRAM calculator.
Similar vision-language modelsLink to this section
- Qwen3.8-Flash-Next - 180B, 2,727,999 downloads/month
- MiniMax-M3.1-Flash - 220.5B, 150,002 downloads/month
- Qwen3.5 122B A10B - 125.1B, 118,592 downloads/month
- Inkling Small - 156B, 80,407 downloads/month
- MiMo-V2.5 - 310.8B, 61,254 downloads/month
PopularityLink to this section
Qwen3.8-Flash-Next-E384-FP4-V2 was downloaded 72,942 times on Hugging Face in the last month (72,942 all-time) and has 2,441 likes.
Run it locallyLink to this section
Weights and usage instructions are on the autotrust/Qwen3.8-Flash-Next-E384-FP4-V2 model card; it loads with vllm.
Local runtimes that fit this model:
- vllm - high-throughput serving on NVIDIA GPUs
From the community
What users are saying
“the data to post training not include Vietnamese, that maybe reuslt in this problem.”
Frequently asked
Quick answers to common questions
How much VRAM does Qwen3.8-Flash-Next-E384-FP4-V2 need?
Qwen3.8-Flash-Next-E384-FP4-V2 with 155B parameters needs approximately 90 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.
Is Qwen3.8-Flash-Next-E384-FP4-V2 better than other autotrust models?
Qwen3.8-Flash-Next-E384-FP4-V2 has 155B parameters with 262,144 context - a strong choice for vision, long-context.
What license is Qwen3.8-Flash-Next-E384-FP4-V2 under?
Qwen3.8-Flash-Next-E384-FP4-V2 is released under the Other license, making it suitable for most commercial and personal projects.
What hardware runs Qwen3.8-Flash-Next-E384-FP4-V2 well?
With 155B parameters, Qwen3.8-Flash-Next-E384-FP4-V2 requires adequate VRAM. High-end GPUs like the RTX 4090 (24GB), RTX 5090 (32GB), or Mac Studio with unified memory are good options. Check our hardware directory for specific recommendations.
What is the best quantization for Qwen3.8-Flash-Next-E384-FP4-V2?
Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~110 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.
How long can Qwen3.8-Flash-Next-E384-FP4-V2's context window handle?
Qwen3.8-Flash-Next-E384-FP4-V2 supports a 262,144-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.
What models compete with Qwen3.8-Flash-Next-E384-FP4-V2?
Qwen3.8-Flash-Next-E384-FP4-V2 competes with other 78B–233B. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.
Nearby options
Similar models and compatible hardware by spec