Model benchmarks

Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram local LLM performance

As of November 2026, Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram runs at up to 151.1 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).

llama.cpp
ShareRedditX

Model size

35B

Peak speed

151.1 tok/s

Average speed

150.8 tok/s

Avg PP

2268.8 tok/s

Min memory

n/a

Max context

65,536 tokens

Avg output / run

34,601 tokens

Avg runtime / run

4m 8s

Avg quality

86.1

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)
Overall85.286.187.0
Agent Workflow88.890.191.4
Code Generation80.282.184.0
Role Play & Narrative80.483.085.6
Research & Analysis88.689.189.5

Performance by hardware and tool

Every hardware/tool/quantization combination Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
AMD Radeon AI PRO R9700llama.cpp—151.1 tok/s150.8 tok/sn/a65,536 tokens86.12

Benchmark runs

All 2 Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Frequently asked questions

Is Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram good for agentic (tool-using) tasks?
In our benchmarks, Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram scores 90.1/100 for agentic workflows. It runs at about 150.8 tok/s, so if you want more speed, swift-1.5-qwen3.8-flash-next-iq3_s is faster (~259.0 tok/s) and still scores well for agentic workflows (88.9/100).
How fast is Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram for local inference?
Across 2 community benchmark runs, Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram reaches up to 151.1 tok/s and averages 150.8 tok/s, with the fastest results on AMD Radeon AI PRO R9700.
Which tools have been used to run Ornith-1.5-35B-A3B-APEX-MTP-I-Quality-MTP-ngram?
Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.