Ornith-1.5-35B-A3B-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2

Qwen family · 36.0B params

1recorded run

108.4best tok/s

0verified

Evidence: 0 sourced runs · 1 awaiting review · 1 independent source · first recorded Aug 21, 2026. What these labels mean.

What the Ornith-1.5-35B-A3B-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2 numbers mean

The fastest result on our board is 108.4 output tok/s on an Intel Arc Pro B70 at GPTQ-Int4 with a 2048-token context. That is about 9.2 ms per generated token — 500 tokens takes roughly 4.6 seconds. Prompt processing on the same record runs at 9072.9 tok/s, so a 2,000-token prompt is read in about 0.2 seconds.

How much VRAM does Ornith-1.5-35B-A3B-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2 need?

Ornith-1.5-35B-A3B-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2 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)

Intel Arc Pro B70108.4

Models with a similar footprint

Model	Params	Q4_K_M weights
Ornith-1.5-35B-A3B	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-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
Intel Arc Pro B70 · 32 GB	Ornith-1.5-35B-A3B-GPTQ-Int4-sym-G128-MTP-BF16-MixedCal-v2	GPTQ-Int4	vllm —	2048	108.4 …	9072.9	—	src