DiggTop ⌕

Godoter-27B

Qwen family · 28.0B params

1recorded run
3.2best tok/s
0verified

Evidence: 0 sourced runs · 1 awaiting review · 1 independent source · first recorded Aug 14, 2026. What these labels mean.

What the Godoter-27B numbers mean

The fastest result on our board is 3.2 output tok/s on an RTX 4070 at Q4_K_M with a 32768-token context. That is about 308.2 ms per generated token — 500 tokens takes roughly 154.1 seconds. Prompt processing on the same record runs at 446.5 tok/s, so a 2,000-token prompt is read in about 4.5 seconds. Peak VRAM was measured at 9.5 GB.

How much VRAM does Godoter-27B need?

Godoter-27B has 28.0B 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.

QuantisationEstimated VRAMSmallest card that fits
Q4_K_M (4-bit)18.9 GB24 GB
Q8_0 (8-bit)35.0 GB48 GB
FP16 (16-bit)65.2 GB80 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)
  1. RTX 40703.2

Models with a similar footprint

ModelParamsQ4_K_M weights
Bonsai-27B28B15.8 GB
Huihui-Qwen3.8-27B-abliterated28B15.8 GB
Qwen3.5-27B28B15.8 GB
Qwen3.6-27B28B15.8 GB
Qwen3.6-27B-i128B15.8 GB
Qwen3.6-27B-MTP28B15.8 GB

Ranked by distance from Godoter-27B's parameter count. Full benchmark board.

All benchmarks Methodology

GPUModelQuantBackendCtx tok/sprompt tok/sVRAM GBSource
RTX 4070 · 12 GBGodoter-27BQ4_K_Mllama.cpp —327683.2 …446.59.5src
↑