Model benchmarks

Nex-N2.5-mini-APEX-Mini local LLM performance

As of September 2026, Nex-N2.5-mini-APEX-Mini runs at up to 89.6 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).

llama.cpp
ShareRedditX

Model size

Unknown

Peak speed

89.6 tok/s

Average speed

88.2 tok/s

Avg PP

924.3 tok/s

Min memory

16.0 GB

Max context

131,072 tokens

Avg output / run

32,777 tokens

Avg runtime / run

6m 22s

Avg quality

81.1

Benchmark runs

2

GPUs tested

1

Quality by task

Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.

TaskP5 (low)AvgP95 (high)
Overall76.681.185.5
Agent Workflow90.491.392.2
Code Generation54.565.877.1
Role Play & Narrative83.785.887.9
Research & Analysis78.081.484.9

Performance by hardware and tool

Every hardware/tool/quantization combination Nex-N2.5-mini-APEX-Mini has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
NVIDIA GeForce RTX 5060llama.cpp—89.6 tok/s88.2 tok/s16.0 GB131,072 tokens81.12

Benchmark runs

All 2 Nex-N2.5-mini-APEX-Mini runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Frequently asked questions

Is Nex-N2.5-mini-APEX-Mini good for coding?
In our benchmarks, Nex-N2.5-mini-APEX-Mini scores 65.8/100 for coding. It runs at about 88.2 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-MLX-oQ4e is faster (~107.9 tok/s) and still scores well for coding (81.5/100).
Is Nex-N2.5-mini-APEX-Mini good for agentic (tool-using) tasks?
In our benchmarks, Nex-N2.5-mini-APEX-Mini scores 91.3/100 for agentic workflows, among the top 2 for agentic workflows on consumer hardware (≤24 GB VRAM). It runs at about 88.2 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-MLX-oQ4e-MTP is faster (~123.7 tok/s) and still scores well for agentic workflows (84.4/100).
How fast is Nex-N2.5-mini-APEX-Mini for local inference?
Across 2 community benchmark runs, Nex-N2.5-mini-APEX-Mini reaches up to 89.6 tok/s and averages 88.2 tok/s, with the fastest results on NVIDIA GeForce RTX 5060.
How much memory does Nex-N2.5-mini-APEX-Mini need?
The leanest observed configuration used about 16.0 GB of memory.
Which tools have been used to run Nex-N2.5-mini-APEX-Mini?
Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.