Ornith-1.5-35B-A3B

Qwen family · 36.0B params

9recorded runs

258.8best tok/s

7verified

Evidence: 7 sourced runs · 2 awaiting review · 1 independent source · first recorded Aug 19, 2026 · last recorded Sep 4, 2026. What these labels mean.

What the Ornith-1.5-35B-A3B numbers mean

The fastest verified result on our board is 258.8 output tok/s on an RTX 5090 at NVFP4 with a 2048-token context. That is about 3.9 ms per generated token — 500 tokens takes roughly 1.9 seconds. Across the 9 records we hold for this model the range is 35.4–258.8 tok/s on 8 GPUs, because quantisation, context length and the engine build all move the number.

How much VRAM does Ornith-1.5-35B-A3B need?

Ornith-1.5-35B-A3B has 36.0B parameters. These estimates use the same formula as the VRAM calculator: weights, plus about 15% for a moderate-context KV cache and runtime overhead, plus a small engine floor.

Quantisation	Estimated VRAM	Smallest card that fits
Q4_K_M (4-bit)	24.1 GB	32 GB
Q8_0 (8-bit)	44.8 GB	64 GB
FP16 (16-bit)	83.6 GB	141 GB

Long contexts and large batches raise the real number — read the smallest card as "possible with a reduced context". Measured throughput is on the benchmark board.

Fastest GPUs for this model (best tok/s)

Models with a similar footprint

Model	Params	Q4_K_M weights
Ornith-1.5-35B-A3B-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2	36B	20.2 GB
Ornith-1.5-35B-A3B-MTP	36B	20.2 GB
Qwen3.6-35B-A3B	36B	20.2 GB
Qwen3.6-35B-A3B-abliterated-v4	36B	20.2 GB
Qwen3.6-35B-A3B-uncensored-heretic-Native-MTP-Preserved	36B	20.2 GB
SIQ-1-35B-B70-Turbo	36B	20.2 GB

Ranked by distance from Ornith-1.5-35B-A3B's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX 5090 · 32 GB	Ornith-1.5-35B-A3B	NVFP4	vllm v0.27.1	2048	258.8 ✓	—	—	src
Intel Arc Pro B70 · 32 GB	Ornith-1.5-35B-A3B	Q4_K_M	llama.cpp 9fee29e9435f + neural.download twelv…	8192	131.5 ✓	—	—	src
CMP 170HX · 64 GB	Ornith-1.5-35B-A3B	Q8_0	llama.cpp 0.3.0-dev+b10705	131338	87.1 ✓	1329.4	—	src
Tesla V100 x2 · 32 GB	Ornith-1.5-35B-A3B	Q4_K_M	llama.cpp 0.1.0-dev (build 10434)	131072	65.0 ✓	305.0	25.2	src
RTX 3080 · 10 GB	Ornith-1.5-35B-A3B	Q4_K_M	llama.cpp b10472	131072	64.4 ✓	596.3	9.6	src
CMP 170HX · 64 GB	Ornith-1.5-35B-A3B	W4A16	vllm 0.27.1	131335	48.6 ✓	5318.3	—	src
Radeon AI Pro R9700 x3 · 32 GB	Ornith-1.5-35B-A3B	BF16	llama.cpp b10679-50f068fff	262144	35.4 ✓	1454.2	—	src
RTX 3060 · 12 GB	Ornith-1.5-35B-A3B	APEX-MTP-I-Compact	llama.cpp —	65536	54.3 …	1341.4	—	src
RTX 4070 · 12 GB	Ornith-1.5-35B-A3B	Q6_K	llama.cpp —	215000	42.7 …	36.0	11.6	src