Local AI Registry

Recipe

qwen3-6-35b-a3b-iq3-xs-rx-9070-xt-16gb-llama-cpp-tp1

qwen3-6-35b-a3b-iq3-xs-rx-9070-xt-16gb-llama-cpp-tp1
JSON APIFind compatible recipes

Observed LocalMaxxing leaderboard run. Evidence for compatibility, not an executable launch contract.

Record

Status
candidate
Source
localmaxxing
Engine
llama.cpp
Engine version
llama-b8892
Accelerators
1
Tensor parallel
1
Context tokens
65,536
Max concurrency
1
chat
unknown
reasoning
unknown
tools
unknown
vision
unknown

Hugging Face model card

Identity

https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF
Repository
unsloth/Qwen3.6-35B-A3B-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/cmofgfg5t000ljs04iip74hzt

Observed source tokens

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

  1. ./llama-server
  2. --threads
  3. 7
  4. --ctx-size
  5. 65536
  6. --keep
  7. 512
  8. --flash-attn
  9. on
  10. --fit
  11. on
  12. --fit-ctx
  13. 65536
  14. --main-gpu
  15. 0
  16. --split-mode
  17. none
  18. --no-mmap
  19. --context-shift
  20. --reasoning
  21. off
  22. --temp
  23. 0.6
  24. --top-p
  25. 0.95
  26. --top-k
  27. 20
  28. --repeat-penalty
  29. 1.05
  30. --min-p
  31. 0.00
  32. --model
  33. /mnt/816d99c9-d5f0-40ae-96fd-b44f7f62faa4/misc/models/Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf
  34. --host
  35. 192.168.1.102
  36. --port
  37. 8080
FlagValue
--threads7
--ctx-size65536
--keep512
--flash-attnon
--fiton
--fit-ctx65536
--main-gpu0
--split-modenone
--reasoningoff
--temp0.6
--top-p0.95
--top-k20
--repeat-penalty1.05
--min-p0.00
--model/mnt/816d99c9-d5f0-40ae-96fd-b44f7f62faa4/misc/models/Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf
--host192.168.1.102
--port8080

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatusSweep
165,536—76360observedqwen3-6-35b-a3b-iq3-xs-rx-9070-xt-16gb-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
rx-9070-xt-16gb
id
qwen3-6-35b-a3b-iq3-xs-rx-9070-xt-16gb-llama-cpp-tp1
model instance id
unsloth-qwen3-6-35b-a3b-gguf--iq3-xs
recipe source
localmaxxing
schema version
local-ai-registry/v1
speed sweep ids
qwen3-6-35b-a3b-iq3-xs-rx-9070-xt-16gb-llama-cpp-tp1-sweep
status
candidate

capabilities

engine

name
llama.cpp
version
llama-b8892

serving

max concurrency
1
max context tokens
65,536
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 9070
localmaxxing · notes
installed rocm 7.2.2 with this guide https://www.fixnum.org/2025-11-27-rocm7-fedora43/ IQ3_XXS gguf, works well with qwen code but had to add a bit of repetition penalty because it got stuck in thinking occasionally
localmaxxing · observed command
./llama-server --threads 7 --ctx-size 65536 --keep 512 --flash-attn on --fit on --fit-ctx 65536 --main-gpu 0 --split-mode none --no-mmap --context-shift --reasoning off --temp 0.6 --top-p 0.95 --top-k 20 --repeat-penalty 1.05 --min-p 0.00 --model /mnt/816d99c9-d5f0-40ae-96fd-b44f7f62faa4/misc/models/Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf --host 192.168.1.102 --port 8080
localmaxxing · run id
cmofgfg5t000ljs04iip74hzt
localmaxxing · tokenized · arguments
./llama-server, --threads, 7, --ctx-size, 65536, --keep, 512, --flash-attn, on, --fit, on, --fit-ctx, 65536, --main-gpu, 0, --split-mode, none, --no-mmap, --context-shift, --reasoning, off, --temp, 0.6, --top-p, 0.95, --top-k, 20, --repeat-penalty, 1.05, --min-p, 0.00, --model, /mnt/816d99c9-d5f0-40ae-96fd-b44f7f62faa4/misc/models/Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf, --host, 192.168.1.102, --port, 8080
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/cmofgfg5t000ljs04iip74hzt ↗