Local AI Registry

Recipe

qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr

qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr
JSON APIFind compatible recipes

Observed 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
8,192
Max concurrency
1
chat
unknown
reasoning
unknown
tools
unknown
vision
unknown

Hugging Face model card

Identity

https://huggingface.co/Jackrong/Qwopus3.6-27B-v2-MTP-GGUF
Repository
Jackrong/Qwopus3.6-27B-v2-MTP-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

Source
https://www.localmaxxing.com/en/runs/cmqjwmi8600s3nv0170ju7hgr

Observed source tokens

Mechanical split of the source command. Unverified against the engine CLI. The registry does not offer Run for this recipe.

  1. -m
  2. /home/ghazni/models/Qwopus3.6-27B-v2-MTP-Q4_K_M.gguf
  3. --host
  4. 0.0.0.0
  5. --port
  6. 8080
  7. -ngl
  8. 99
  9. --no-mmap
  10. -c
  11. 150000
  12. -fa
  13. on
  14. -ctk
  15. q8_0
  16. -ctv
  17. q8_0
  18. --reasoning
  19. on
  20. -np
  21. 1
  22. --spec-type
  23. draft-mtp
  24. --spec-draft-n-max
  25. 2
  26. -fit
  27. off
  28. Restart=on-failure
  29. RestartSec=10
  30. StandardOutput=append:/home/ghazni/llama-server.log
  31. StandardError=append:/home/ghazni/llama-server.log
  32. TimeoutStartSec=300
  33. [Install]
  34. WantedBy=multi-user.target
FlagValue
-m/home/ghazni/models/Qwopus3.6-27B-v2-MTP-Q4_K_M.gguf
--host0.0.0.0
--port8080
-ngl99
-c150000
-faon
-ctkq8_0
-ctvq8_0
--reasoningon
-np1
--spec-typedraft-mtp
--spec-draft-n-max2
-fitoff

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatusSweep
18,192664.9501,452.9observedqwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr-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
rx-7900-xtx-24gb
id
qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr
model instance id
jackrong-qwopus3-6-27b-v2-mtp-gguf--q4-k-m
recipe source
localmaxxing
schema version
local-ai-registry/v1
speed sweep ids
qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr-sweep
status
candidate

capabilities

engine

name
llama.cpp

serving

max concurrency
1
max context tokens
8,192
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

metadata

localmaxxing · backend
rocm
localmaxxing · hardware label
RX 7900 XTX
localmaxxing · notes
Benchmark via localmaxx-bench CLI. MTP speculative decoding (draft-mtp, n_max=2). Q8_0 KV cache. ROCm backend on Ubuntu 26.04. Flags auto-detected from service file.
localmaxxing · observed command
-m /home/ghazni/models/Qwopus3.6-27B-v2-MTP-Q4_K_M.gguf --host 0.0.0.0 --port 8080 -ngl 99 --no-mmap -c 150000 -fa on -ctk q8_0 -ctv q8_0 --reasoning on -np 1 --spec-type draft-mtp --spec-draft-n-max 2 -fit off Restart=on-failure RestartSec=10 StandardOutput=append:/home/ghazni/llama-server.log StandardError=append:/home/ghazni/llama-server.log TimeoutStartSec=300 '[Install]' WantedBy=multi-user.target
localmaxxing · run id
cmqjwmi8600s3nv0170ju7hgr
localmaxxing · tokenized · arguments
-m, /home/ghazni/models/Qwopus3.6-27B-v2-MTP-Q4_K_M.gguf, --host, 0.0.0.0, --port, 8080, -ngl, 99, --no-mmap, -c, 150000, -fa, on, -ctk, q8_0, -ctv, q8_0, --reasoning, on, -np, 1, --spec-type, draft-mtp, --spec-draft-n-max, 2, -fit, off, Restart=on-failure, RestartSec=10, StandardOutput=append:/home/ghazni/llama-server.log, StandardError=append:/home/ghazni/llama-server.log, TimeoutStartSec=300, [Install], WantedBy=multi-user.target
localmaxxing · tokenized · fidelity
faithful

provenance

captured at
2026-08-30T09:10:02Z

sources

captured atkindurl
2026-08-30T09:10:02Znormalized-recipewww.localmaxxing.com/en/runs/cmqjwmi8600s3nv0170ju7hgr ↗