How much VRAM does Qwen3.8 Flash-Next need?
Qwen3.8 Flash-Next needs 102 GB for its weights at 4-bit, 103 GB with an 8K context and 103 GB with 32K. At 8-bit, 179 GB.
- Parameters
- 180B
- Active per token
- 6B
- Longest context
- 256K
- Released
- Aug 2026
- License
- Qwen Community 1.0
- Strongest at
- Chat, Coding, Agents
Memory at each precision
| Precision | Weights | With 8K | With 32K | With 128K |
|---|---|---|---|---|
| 16-bit | 335 GB | 337 GB | 337 GB | 339 GB |
| 8-bit | 178 GB | 179 GB | 180 GB | 182 GB |
| 6-bit | 137 GB | 139 GB | 139 GB | 142 GB |
| 5-bit | 119 GB | 121 GB | 121 GB | 123 GB |
| 4-bit | 102 GB | 103 GB | 103 GB | 106 GB |
| 3-bit | 81.9 GB | 83.2 GB | 83.8 GB | 86.0 GB |
| 2-bit | 62.9 GB | 64.2 GB | 64.7 GB | 67.0 GB |
| 1.6-bit | 34.6 GB | 35.9 GB | 36.4 GB | 38.7 GB |
One conversation, 16-bit context cache, runtime overhead included.
What runs it
| Machine | Best fit | Context | Speed | First reply |
|---|---|---|---|---|
| RTX 3060 12 GB | Too big | |||
| RTX 4060 Ti 16 GB | Too big | |||
| RTX 3090 | Too big | |||
| RTX 4090 | Too big | |||
| RTX 5090 | Too big | |||
| 2× RTX 3090 | Too big | |||
| RTX PRO 6000 Blackwell | 3-bit | 256K | 188 tok/s | 0.3 s |
| M4 Max 128 GB | 3-bit | 256K | 65 tok/s | 5.7 s |
| M5 Max 128 GB | 3-bit | 256K | 70 tok/s | 1.6 s |
| Ryzen AI Max+ 395 128 GB | 3-bit | 256K | 41 tok/s | 3.3 s |
| DGX Spark 128 GB | 4-bit | 256K | 44 tok/s | 2.3 s |
| M3 Ultra 256 GB | 8-bit | 136K | 53 tok/s | 3.2 s |
| M5 Ultra 512 GB | 8-bit | 256K | 66 tok/s | 1.0 s |
| H100 SXM | 2-bit | 256K | 257 tok/s | 0.2 s |
| DGX Station GB300 | 8-bit | 256K | 239 tok/s | 0.2 s |
| 8× H200 NVL | 8-bit | 256K | 208 tok/s | 0.1 s |
Best fit is the highest quality, up to 8-bit, that leaves 16K of context. First reply is the wait for a 2K prompt. Speeds are rough, plus or minus 25%.
How to run it
- Ollama
ollama run qwen3.8-flash-next:125b-a6b-q4_K_M- llama.cpp
llama-server -hf unsloth/Qwen3.8-Flash-Next-GGUF- vLLM
vllm serve Qwen/Qwen3.8-Flash-Next- MLX (Mac)
mlx_lm.chat --model mlx-community/Qwen3.8-Flash-Next-4bit
Questions
How much VRAM does Qwen3.8 Flash-Next need?
Qwen3.8 Flash-Next needs 102 GB for its weights at 4-bit, 103 GB with an 8K context and 103 GB with 32K. At 8-bit, 179 GB.
Can Qwen3.8 Flash-Next run on an RTX 4090?
Not on one card. The smallest machine here that runs it is the RTX PRO 6000 Blackwell.