Open-weight model
Qwen3.8-Flash-Next-oQ6e-fp16-mtp
by Scott Lowry scottlowry/Qwen3.8-Flash-Next-oQ6e-fp16-mtp
Qwen3.8-Flash-Next-oQ6e-fp16-mtp is an open-weight model from Scott Lowry. It has 180B parameters and a 262,144-token context. At 16-bit it needs about 432 GB of GPU memory, which fits on 2x MI325X from $4.00 an hour; at 4-bit, 108 GB on 1x MI300X from $1.85, at the lowest prices in the SAVRN Index.
This model was quantized using oQ (oMLX v0.7.1.dev1) mixed-precision quantization.
Runs On
What it takes to serve Qwen3.8-Flash-Next-oQ6e-fp16-mtp (180B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 360.0 GB | 432.0 GB | 2x MI325X (256 GB) Vultr |
$4.00 | 2x MI355X $5.18 · 3x MI300X $5.55 |
| 8-bit | 180.0 GB | 216.0 GB | 1x MI325X (256 GB) Vultr |
$2.00 | 1x MI355X $2.59 · 2x MI300X $3.70 |
| 4-bit | 90.0 GB | 108.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x MI325X $2.00 · 1x MI355X $2.59 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 23, 2026.
Qwen3.8-Flash-Next-oQ6e-fp16-mtp on every accelerator the SAVRN Index prices, at every precision
Model Card
This model was quantized using oQ (oMLX v0.7.1.dev1) mixed-precision quantization.
Excerpt from the card by Scott Lowry.
Configuration
- Architecture
- Qwen4ExpForConditionalGeneration
- Context length (tokens)
- 262,144
- Layers
- 48
- Hidden size
- 2,560
- Attention heads
- 24
- Key/value heads
- 2
- Head dimension
- 256
- Vocabulary size
- 248,320
- Experts
- 512
- Experts active per token
- 10
- Model type
- qwen4_exp
Identity and Version
- Repository
- scottlowry/Qwen3.8-Flash-Next-oQ6e-fp16-mtp
- Publisher
- Scott Lowry
- Task
- Not stated by the source
- Modality
- Other
- Library
- mlx
- Parameters
- 180B parameters
- Languages
- mlx, oq
- Revision
- 88c8aaf7c21a1e699b5949dd622b875780ae5d6c
- First published
- 2026-10-14
- Last updated
- 2026-10-14
Files and Weights
41 files, 151.5 GB in total. The weights are 29 files totalling 151.4 GB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00029.safetensors | Weights | 5.1 GB | 17d3669d6b81 |
| model-00002-of-00029.safetensors | Weights | 5.3 GB | 9e7b33576f04 |
| model-00003-of-00029.safetensors | Weights | 5.3 GB | e85ebe6710b3 |
| model-00004-of-00029.safetensors | Weights | 5.3 GB | f3def38c5f17 |
| model-00005-of-00029.safetensors | Weights | 5.3 GB | 2e6d934d7b58 |
| model-00006-of-00029.safetensors | Weights | 5.3 GB | 9f5ad45dad9e |
| model-00007-of-00029.safetensors | Weights | 5.3 GB | 712f20154a2d |
| model-00008-of-00029.safetensors | Weights | 5.3 GB | 6f6d95269cef |
| model-00009-of-00029.safetensors | Weights | 5.0 GB | d265f844971b |
| model-00010-of-00029.safetensors | Weights | 5.1 GB | a649f57441fd |
| model-00011-of-00029.safetensors | Weights | 5.5 GB | 0fc01e9ee1ee |
| model-00012-of-00029.safetensors | Weights | 5.5 GB | 3669b9a64b7d |
| model-00013-of-00029.safetensors | Weights | 5.5 GB | d1ea06789a2e |
| model-00014-of-00029.safetensors | Weights | 5.5 GB | 1bff8a7c3115 |
| model-00015-of-00029.safetensors | Weights | 5.5 GB | 831d6fbe0d33 |
| model-00016-of-00029.safetensors | Weights | 5.5 GB | 7cbc2bede7e1 |
| model-00017-of-00029.safetensors | Weights | 5.5 GB | e411f3f69bd6 |
| model-00018-of-00029.safetensors | Weights | 5.5 GB | 11537d1ac6d3 |
| model-00019-of-00029.safetensors | Weights | 5.5 GB | 394308a48fb8 |
| model-00020-of-00029.safetensors | Weights | 5.5 GB | 1da20555eeeb |
| model-00021-of-00029.safetensors | Weights | 5.5 GB | 93b74ad63c71 |
| model-00022-of-00029.safetensors | Weights | 5.5 GB | c83cb92a95f5 |
| model-00023-of-00029.safetensors | Weights | 5.5 GB | 629019e6a971 |
| model-00024-of-00029.safetensors | Weights | 5.5 GB | 9d47185b2f95 |
| model-00025-of-00029.safetensors | Weights | 5.5 GB | 51c562cabf6a |
| model-00026-of-00029.safetensors | Weights | 5.5 GB | ae10469884aa |
| model-00027-of-00029.safetensors | Weights | 5.5 GB | 9c531973cdef |
| model-00028-of-00029.safetensors | Weights | 5.5 GB | dbf38d77d724 |
| model-00029-of-00029.safetensors | Weights | 1.4 GB | dd4cf7f138ba |
| config.json | Configuration | 134.3 KB | — |
| generation_config.json | Configuration | 202 B | — |
| model.safetensors.index.json | Configuration | 405.2 KB | — |
| oq_imatrix_report.json | Configuration | 68.7 KB | — |
| preprocessor_config.json | Configuration | 390 B | — |
| README.md | Documentation | 328 B | — |
| chat_template.jinja | Other | 9.0 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 3.4 MB | — |
| tokenizer.json | Tokenizer | 12.8 MB | 0997f410c57a |
| tokenizer_config.json | Tokenizer | 17.9 KB | — |
| vocab.json | Tokenizer | 6.7 MB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 151.4 GB
Released by Scott Lowry through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 151.4 GB |
| 16-bit | 360.0 GB |
| 8-bit | 180.0 GB |
| 4-bit | 90.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About Qwen3.8-Flash-Next-oQ6e-fp16-mtp
How much GPU memory does Qwen3.8-Flash-Next-oQ6e-fp16-mtp need?
About 432 GB at 16-bit and 108 GB at 4-bit: the weights (180B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Qwen3.8-Flash-Next-oQ6e-fp16-mtp on?
At 16-bit, 2x MI325X from $4.00 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
What is Qwen3.8-Flash-Next-oQ6e-fp16-mtp's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.