Recipe

ling-3-0-tiny-q8-0-rtx-3090-24gb-llama-cpp-tp2

ling-3-0-tiny-q8-0-rtx-3090-24gb-llama-cpp-tp2

JSON API Find compatible recipes

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

Record

Status: candidate
Source: localmaxxing
Engine: llama.cpp
Accelerators: 2
Tensor parallel: 2
Context tokens: 2,048
Max concurrency: 2,048
chat: unknown
reasoning: unknown
tools: unknown
vision: unknown

Hugging Face model card

Identity

https://huggingface.co/Mike0021/Ling-3.0-tiny-GGUF

Repository: Mike0021/Ling-3.0-tiny-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/cmsoz8tkw00gfmp01qy5xymx7

Observed source tokens

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

/home/lotto/llama.cpp-ling3/build/bin/llama-bench
--offline
-m
/home/lotto/models/Ling-3.0-tiny-GGUF/Ling-3.0-tiny-Q8_0.gguf
-p
2048
-n
256
-b
2048
-ub
512
-ngl
999
-dev
CUDA0,CUDA1
-sm
layer
-ts
1/1
-fa
on
-r
5
-o
json

Flag	Value
`-m`	`/home/lotto/models/Ling-3.0-tiny-GGUF/Ling-3.0-tiny-Q8_0.gguf`
`-p`	`2048`
`-n`	`256`
`-b`	`2048`
`-ub`	`512`
`-ngl`	`999`
`-dev`	`CUDA0,CUDA1`
`-sm`	`layer`
`-ts`	`1/1`
`-fa`	`on`
`-r`	`5`
`-o`	`json`

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status	Sweep
2048	2,048	9,243.2	232.8	47.8	observed	ling-3-0-tiny-q8-0-rtx-3090-24gb-llama-cpp-tp2-sweep

Remaining fields

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

hardware count: 2
hardware id: rtx-3090-24gb
id: ling-3-0-tiny-q8-0-rtx-3090-24gb-llama-cpp-tp2
model instance id: mike0021-ling-3-0-tiny-gguf--q8-0
recipe source: localmaxxing
schema version: local-ai-registry/v1
speed sweep ids: ling-3-0-tiny-q8-0-rtx-3090-24gb-llama-cpp-tp2-sweep
status: candidate

capabilities

engine

name: llama.cpp

serving

max concurrency: 2,048
max context tokens: 2,048
tensor parallel: 2

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
localmaxxing · notes: Q8_0 on pinned llama.cpp PR 26608 commit d8d862521, CUDA, 2x RTX 3090, all layers GPU-offloaded with 1:1 layer split, 131072 server context, flash attention. Prefill and decode are llama-bench pp2048/tg256 means over 5 repetitions after default warmup: 9243.21531 prompt tok/s and 232.774459 output tok/s. TTFT is median of 5 streaming endpoint requests after 2 warmups with prompt caching explicitly disabled: 47.77 ms (range 47.71-47.97 ms). The endpoint TTFT prompt was 42 actual tokens. Remote semantic decode was 224.4 tok/s median; llama-bench sustained decode is submitted as tokSOut.
localmaxxing · observed command: /home/lotto/llama.cpp-ling3/build/bin/llama-bench --offline -m /home/lotto/models/Ling-3.0-tiny-GGUF/Ling-3.0-tiny-Q8_0.gguf -p 2048 -n 256 -b 2048 -ub 512 -ngl 999 -dev CUDA0,CUDA1 -sm layer -ts 1/1 -fa on -r 5 -o json
localmaxxing · run id: cmsoz8tkw00gfmp01qy5xymx7
localmaxxing · tokenized · arguments: /home/lotto/llama.cpp-ling3/build/bin/llama-bench, --offline, -m, /home/lotto/models/Ling-3.0-tiny-GGUF/Ling-3.0-tiny-Q8_0.gguf, -p, 2048, -n, 256, -b, 2048, -ub, 512, -ngl, 999, -dev, CUDA0,CUDA1, -sm, layer, -ts, 1/1, -fa, on, -r, 5, -o, json
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/cmsoz8tkw00gfmp01qy5xymx7 ↗