Glyd

What runs on a 16 GB GPU, exactly

RTX 4080, 4080 SUPER, 5080, 5070 Ti and 4060 Ti 16 GB. Open models in their original precision, no quantization, with and without Glyd.

Measured on our RTX 4080 SUPER
47.2tokens/s, Qwen3-8B, one sequence

Its 16.38 GB of bf16 leave no room to run on this card. Glyd holds it in 11.15 GB, every weight exact. DFloat11 at the same size: 13.8 tokens/s.

Runs only with Glyd

bf16 does not fit; Glyd does, bit for bit
Modelbf16 needsGlyd needsWith Glyd
Mistral 7B v0.317.2 GB12.4 GBRoom for about 44K tokens of context with Glyd
Llama 3.1 8B18.7 GB13.5 GBRoom for about 36K tokens of context with Glyd
Qwen3 8B19.2 GB14.0 GBRoom for about 30K tokens of context with Glyd

Runs either way

Glyd leaves room for a longer context or more users
Modelbf16 needsGlyd needsThe difference
SmolLM3 3B8.4 GB6.3 GBContext room 128K → 155K tokens with Glyd
Llama 3.2 3B9.0 GB6.9 GBContext room 80K → 98K tokens with Glyd
Qwen3 4B 250710.9 GB8.3 GBContext room 51K → 69K tokens with Glyd

Too big for 16 GB

Even compressed
Modelbf16 needsGlyd needsWhat it takes
Gemma 3 12B26.9 GB18.8 GBNeeds 18.8 GB with Glyd: try 24 GB
Phi-4 14B32.6 GB23.0 GBNeeds 23.0 GB with Glyd: try 24 GB
R1 Distill Qwen 14B32.8 GB23.3 GBNeeds 23.3 GB with Glyd: try 24 GB
Mistral Small 3.2 24B51.0 GB35.1 GBNeeds 35.1 GB with Glyd: try 48 GB

Speed on this GPU

RTX 4080 SUPER, Qwen2.5-7B, GPU time a token. Lower is better.

Sequences a stepbf16GlydAgainst bf16
1 sequence21.97 ms16.52 ms25% less time
8 sequences22.85 ms17.37 ms24% less time
32 sequences26.19 ms18.80 ms28% less time
64 sequences27.67 ms25.14 ms9% less time

Questions about 16 GB GPUs

Can I run an 8B model on a 16 GB GPU without quantizing it?

Yes, with Glyd. Llama 3.1 8B and Qwen3 8B take about 16 GB in bf16, which leaves no room to run; Glyd holds them in about 11 GB with every weight exact.

Is it slower than bf16?

On this card it is faster: 9 to 28% less GPU time a token for Qwen2.5-7B, from 1 to 64 sequences a step.

How is this different from a 4-bit GGUF?

A 4-bit model is smaller, about 5 GB for an 8B model, but its weights are rounded and its answers change. Glyd gives you the original model.

Updated Sep 27, 2026. A model fits when its weights, an 8K-token KV cache and 1.5 GB for the runtime fit in the memory nvidia-smi reports: 16,376 MiB on a 16 GB card.