Recipe

kimi-k2-5-iq1-s-rtx-3090-24gb-llama-cpp-tp8

kimi-k2-5-iq1-s-rtx-3090-24gb-llama-cpp-tp8

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

Record

Status
candidate
Source
localmaxxing
Engine
llama.cpp
Engine version
8178 (3e6ab244a)
Accelerators
8
Tensor parallel
8
Context tokens
4,096
chat
unknown
reasoning
unknown
tools
unknown
vision
unknown

Hugging Face model card

Identity

https://huggingface.co/bartowski/moonshotai_Kimi-K2.5-GGUF
Repository
bartowski/moonshotai_Kimi-K2.5-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/cmowla4rq004io201otfp0rdo

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. /data/models/kimi-k25-gguf/bartowski-moonshotai-kimi-k25-iq1s/moonshotai_Kimi-K2.5-IQ1_S-00001-of-00006.gguf
  4. -c
  5. 4096
  6. -b
  7. 512
  8. -ub
  9. 128
  10. -ngl
  11. 48
  12. --split-mode
  13. layer
  14. --tensor-split
  15. 1,1,1,1,1,1,1,1
  16. -fa
  17. -ctk
  18. q4_0
  19. -ctv
  20. q4_0
FlagValue
-m/data/models/kimi-k25-gguf/bartowski-moonshotai-kimi-k25-iq1s/moonshotai_Kimi-K2.5-IQ1_S-00001-of-00006.gguf
-c4096
-b512
-ub128
-ngl48
--split-modelayer
--tensor-split1,1,1,1,1,1,1,1
-ctkq4_0
-ctvq4_0

Environment

VariableValue
CUDA_VISIBLE_DEVICES2,3,4,5,6,7,8,9

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatusSweep
14,0965.5142observedkimi-k2-5-iq1-s-rtx-3090-24gb-llama-cpp-tp8-sweep

Remaining fields

Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.

hardware count
8
hardware id
rtx-3090-24gb
id
kimi-k2-5-iq1-s-rtx-3090-24gb-llama-cpp-tp8
model instance id
bartowski-moonshotai-kimi-k2-5-gguf--iq1-s
recipe source
localmaxxing
schema version
local-ai-registry/v1
speed sweep ids
kimi-k2-5-iq1-s-rtx-3090-24gb-llama-cpp-tp8-sweep
status
candidate

capabilities

engine

name
llama.cpp
version
8178 (3e6ab244a)

serving

max context tokens
4,096
tensor parallel
8
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-not-evidenced
serving.max concurrency · state unknown

metadata

localmaxxing · backend
cuda
localmaxxing · batch size
1
localmaxxing · hardware label
RTX 3090
localmaxxing · notes
Kimi-K2.5 IQ1_S GGUF feasibility row on 8x RTX 3090. Full offload did not fit; first stable route used n_gpu_layers=48 with substantial CPU offload, so this is not a speed-match row. 3 measured 512-token runs after 1 warmup; wall tok/s varied 6.21, 6.44, 3.94. Raw log: /home/mattsav/Documents/data/glm51_8x3090_20260507/kimi_k25_iq1s_8x_ngl48_b512_bench512.log
localmaxxing · observed command
CUDA_VISIBLE_DEVICES=2,3,4,5,6,7,8,9 llama-server -m /data/models/kimi-k25-gguf/bartowski-moonshotai-kimi-k25-iq1s/moonshotai_Kimi-K2.5-IQ1_S-00001-of-00006.gguf -c 4096 -b 512 -ub 128 -ngl 48 --split-mode layer --tensor-split 1,1,1,1,1,1,1,1 -fa -ctk q4_0 -ctv q4_0
localmaxxing · run id
cmowla4rq004io201otfp0rdo
localmaxxing · tokenized · arguments
llama-server, -m, /data/models/kimi-k25-gguf/bartowski-moonshotai-kimi-k25-iq1s/moonshotai_Kimi-K2.5-IQ1_S-00001-of-00006.gguf, -c, 4096, -b, 512, -ub, 128, -ngl, 48, --split-mode, layer, --tensor-split, 1,1,1,1,1,1,1,1, -fa, -ctk, q4_0, -ctv, q4_0
localmaxxing · tokenized · environment · CUDA VISIBLE DEVICES
2,3,4,5,6,7,8,9
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/cmowla4rq004io201otfp0rdo