Qwen3.8-27B UD-Q6_K_M on the Radeon AI PRO R9700 32GB (Linux)
Candidate
Candidate Linux single-card configuration for Radeon AI PRO R9700 32 GB; no published fit or performance result.
What will be tested
| Card | 1 × AMD Radeon AI PRO R9700 32GB |
|---|---|
| Operating system | Linux |
| Backend | Not yet selected |
| Model | Qwen3.8-27B (Apache-2.0) |
| Model file | Unsloth GGUF conversion, UD-Q6_K_M |
| Context | 32,768 tokens |
| Concurrent requests | 1 (one request at a time) |
| Placement | Full GPU |
| Reviewed | Not yet |
| Sources | Upstream model · GGUF file page |
What to expect
No published results yet. Measured results appear here only after testing and review.
Install
Candidate — no installer yet.
Overview
This Linux candidate pairs the Radeon AI PRO R9700 32 GB with the pinned model and quantization named above at the catalog's 32,768-token context. The 32 GB workstation card is paired with a Q6 27B artifact at a 32K catalog context. A separate Q8 candidate exists; whether either profile has adequate free memory and useful response behavior requires exact-card evidence. It is a test plan, not a claim of load, full-GPU placement, answer quality or speed.
Requirements
The current Linux path is a private, supervised source-checkout workflow for one recognized Vulkan GPU and a loopback server. It requires explicit artifact consent. No physical Linux AMD serving outcome is qualified. Check the card's currently reported free GPU memory, not just the number printed on its box. A longer prompt, more active requests and the desktop can change the memory budget. Keep the exact driver and operating-system build with any later test.
The pinned GGUF is 23,088,409,504 bytes (23.09 GB decimal). Keep more free local SSD space than the artifact itself for verified acquisition and cache maintenance; no fixed extra margin or host-RAM minimum has been qualified. Review the exact model license before acquisition. A planned trial must record the actual free-memory budget, selected device, loaded context, host and GPU buffers, delivered output and representative answers. The catalog context is an intended setting, not a demonstrated maximum.
Limitations
- The exact-card load, context, GPU/host buffers, response quality and speed have no published qualification result.
- The Linux path is a private source-checkout single-card workflow, not physically qualified on this card or released to consumers.
- This candidate requests full-GPU placement; it is not a verified physical-residency or all-operations-on-GPU claim.
