Local AI Registry

Recipe

qwen3-8b-base-q4-k-m-rtx-3090-ti-24gb-llama-cpp-tp1

qwen3-8b-base-q4-k-m-rtx-3090-ti-24gb-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
b1-6217b49 (CUDA)
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/Qwen/Qwen3-8B
Repository
Qwen/Qwen3-8B
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/cmogsu1xc000mjv04iforumvd

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. -m
  3. Qwen3-8B-Q4_K_M.gguf
  4. --alias
  5. qwen3-8b-q4
  6. -ngl
  7. 999
  8. -fa
  9. 1
  10. --no-mmap
  11. -c
  12. 8192
  13. -ctk
  14. q8_0
  15. -ctv
  16. q8_0
  17. -b
  18. 2048
  19. -ub
  20. 2048
  21. --metrics
  22. --jinja
  23. --temp
  24. 1.0
  25. --top-p
  26. 0.95
  27. --top-k
  28. 20
  29. --min-p
  30. 0.0
  31. --host
  32. 127.0.0.1
  33. --port
  34. 8003
FlagValue
-mQwen3-8B-Q4_K_M.gguf
--aliasqwen3-8b-q4
-ngl999
-fa1
-c8192
-ctkq8_0
-ctvq8_0
-b2048
-ub2048
--temp1.0
--top-p0.95
--top-k20
--min-p0.0
--host127.0.0.1
--port8003

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatusSweep
18,192—133122.6observedqwen3-8b-base-q4-k-m-rtx-3090-ti-24gb-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
rtx-3090-ti-24gb
id
qwen3-8b-base-q4-k-m-rtx-3090-ti-24gb-llama-cpp-tp1
model instance id
qwen-qwen3-8b--q4-k-m
recipe source
localmaxxing
schema version
local-ai-registry/v1
speed sweep ids
qwen3-8b-base-q4-k-m-rtx-3090-ti-24gb-llama-cpp-tp1-sweep
status
candidate

capabilities

engine

name
llama.cpp
version
b1-6217b49 (CUDA)

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
cuda
localmaxxing · hardware label
RTX 3090 Ti
localmaxxing · notes
Qwen 3-8B Q4_K_M via llama-server on single RTX 3090 Ti (slot 1 Gen4 x16). Bench: vllm bench serve openai-chat random 512/512 num-prompts 20 concurrency 1 --ignore-eos. TPOT 7.29 ms, ITL 7.34 ms — 8B Q4 is HBM-bandwidth-saturated on 3090 Ti at ~135 t/s decode. KV q8_0, ngl 999, fa on, no spec decode. ~5 GB on disk, ~7 GB peak VRAM. **+51% output throughput vs M3 Ultra entry (88.25 t/s) on the same model+quant** — Ti's 1008 GB/s HBM2 + tensor cores beat M3 Ultra's unified-memory bandwidth on dense 8B inference.
localmaxxing · observed command
llama-server -m Qwen3-8B-Q4_K_M.gguf --alias qwen3-8b-q4 -ngl 999 -fa 1 --no-mmap -c 8192 -ctk q8_0 -ctv q8_0 -b 2048 -ub 2048 --metrics --jinja --temp 1.0 --top-p 0.95 --top-k 20 --min-p 0.0 --host 127.0.0.1 --port 8003
localmaxxing · run id
cmogsu1xc000mjv04iforumvd
localmaxxing · tokenized · arguments
llama-server, -m, Qwen3-8B-Q4_K_M.gguf, --alias, qwen3-8b-q4, -ngl, 999, -fa, 1, --no-mmap, -c, 8192, -ctk, q8_0, -ctv, q8_0, -b, 2048, -ub, 2048, --metrics, --jinja, --temp, 1.0, --top-p, 0.95, --top-k, 20, --min-p, 0.0, --host, 127.0.0.1, --port, 8003
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/cmogsu1xc000mjv04iforumvd ↗