Recipe
qwen3-8-27b-q6-k-xl-radeon-ai-pro-r9700-32gb-llama-cpp-tp1
qwen3-8-27b-q6-k-xl-radeon-ai-pro-r9700-32gb-llama-cpp-tp1Observed LocalMaxxing leaderboard run. Evidence for compatibility, not an executable launch contract.
Record
- Status
- candidate
- Source
- localmaxxing
- Engine
- llama.cpp
- Accelerators
- 1
- Tensor parallel
- 1
- Context tokens
- 32,768
- Max concurrency
- 1
- chat
- unknown
- reasoning
- unknown
- tools
- unknown
- vision
- unknown
Hugging Face model card
Identity
https://huggingface.co/unsloth/Qwen3.8-27B-GGUF- Repository
- unsloth/Qwen3.8-27B-GGUF
- Status
- known
- Link type
- Exact Hub repository
Public Hugging Face repository confirmed by the Hub API.
Candidate: useful compatibility or speed evidence. The registry does not offer Run until promotion requirements are met.
Observed configuration
llama.cpp
Evidence only · candidate · reference
Candidate evidence — not a Run contract
Observed source tokens
Mechanical split of the source command. Unverified against the engine CLI. The registry does not offer Run for this recipe.
llama-server-m<model>--ctx-size32768--parallel1--flash-attnon--cache-type-kq8_0--cache-type-vq8_0--spec-typedraft-mtp--spec-draft-n-max2--no-mmap--deviceVulkan0
| Flag | Value |
|---|---|
-m | <model> |
--ctx-size | 32768 |
--parallel | 1 |
--flash-attn | on |
--cache-type-k | q8_0 |
--cache-type-v | q8_0 |
--spec-type | draft-mtp |
--spec-draft-n-max | 2 |
--device | Vulkan0 |
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status | Sweep |
|---|---|---|---|---|---|---|
| 1 | 32,768 | 801 | 34.3 | 639.2 | observed | qwen3-8-27b-q6-k-xl-radeon-ai-pro-r9700-32gb-llama-cpp-tp1-sweep |
Remaining fields
Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.
- hardware count
- 1
- hardware id
- radeon-ai-pro-r9700-32gb
- id
- qwen3-8-27b-q6-k-xl-radeon-ai-pro-r9700-32gb-llama-cpp-tp1
- model instance id
- unsloth-qwen3-8-27b-gguf--q6-k-xl
- recipe source
- localmaxxing
- schema version
- local-ai-registry/v1
- speed sweep ids
- qwen3-8-27b-q6-k-xl-radeon-ai-pro-r9700-32gb-llama-cpp-tp1-sweep
- status
- candidate
capabilities
engine
- name
- llama.cpp
serving
- max concurrency
- 1
- max context tokens
- 32,768
- tensor parallel
- 1
Provenance & metadata (3)
facts
- capabilities.chat · provenance · captured at
- 2026-08-30T09:10:02Z
capabilities.chat · reason capability-not-verified
capabilities.chat · state unknown
- capabilities.reasoning · provenance · captured at
- 2026-08-30T09:10:02Z
capabilities.reasoning · reason capability-not-verified
capabilities.reasoning · state unknown
- capabilities.tools · provenance · captured at
- 2026-08-30T09:10:02Z
capabilities.tools · reason capability-not-verified
capabilities.tools · state unknown
- capabilities.vision · provenance · captured at
- 2026-08-30T09:10:02Z
capabilities.vision · reason capability-not-verified
capabilities.vision · state unknown
- engine.graph mode · provenance · captured at
- 2026-08-30T09:10:02Z
engine.graph mode · reason runtime-detail-not-published
engine.graph mode · state unknown
- serving.kv cache tokens · provenance · captured at
- 2026-08-30T09:10:02Z
serving.kv cache tokens · reason kv-cache-capacity-not-published
serving.kv cache tokens · state unknown
- serving.max concurrency · provenance · captured at
- 2026-08-31T23:03:15Z
serving.max concurrency · reason server-capacity-derived-from-source-evidence
serving.max concurrency · state known
metadata
- localmaxxing · backend
- vulkan
- localmaxxing · batch size
- 1
- localmaxxing · hardware label
- Radeon AI Pro R9700
- localmaxxing · notes
- MTP=2 speculative decoding (--spec-type draft-mtp --spec-draft-n-max 2 --no-mmap), measured via llama-server on AMD Radeon AI PRO R9700 (Vulkan). Separate side-by-side run vs the standard llama-bench submission.
- localmaxxing · observed command
- llama-server -m <model> --ctx-size 32768 --parallel 1 --flash-attn on --cache-type-k q8_0 --cache-type-v q8_0 --spec-type draft-mtp --spec-draft-n-max 2 --no-mmap --device Vulkan0
- localmaxxing · run id
- cmsuyv0h206wims01hsunwdxw
- localmaxxing · tokenized · arguments
- llama-server, -m, <model>, --ctx-size, 32768, --parallel, 1, --flash-attn, on, --cache-type-k, q8_0, --cache-type-v, q8_0, --spec-type, draft-mtp, --spec-draft-n-max, 2, --no-mmap, --device, Vulkan0
- localmaxxing · tokenized · fidelity
- faithful
provenance
- captured at
- 2026-08-30T09:10:02Z
sources
| captured at | kind | url |
|---|---|---|
| 2026-08-30T09:10:02Z | normalized-recipe | www.localmaxxing.com/en/runs/cmsuyv0h206wims01hsunwdxw ↗ |