selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

fastino logo

Can the Instinct MI50 32GB run GLiDE-27B?

GLiDE-27B·26.1B·Instinct MI50 32GB·32 GB VRAM

Saved in this browser, used on every model page.
Yes

Yes. The Instinct MI50 32GB runs GLiDE-27B fully in VRAM at Q8_0 (about 30 GB of 32 GB), at roughly 20 tokens/s.

best quant
Q8_0
needs
30 GB
usable
32 GB
est. speed
~20 tok/s

Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the GLiDE-27B page →

Every quantisation of GLiDE-27B on the Instinct MI50 32GB

QuantQualityMemoryFitSpeed
Q2_K Heavy loss11 GBfits~66 tok/s
Q3_K_M Noticeable loss15 GBfits~44 tok/s
Q4_K_M Small loss19 GBfits~36 tok/s
Q5_K_M Near-lossless21 GBfits~30 tok/s
Q6_K Near-lossless24 GBfits~26 tok/s
Q8_0 pickLossless30 GBfits~20 tok/s
FP16 / BF16 Lossless54 GBoffload~1.4 tok/s

Speed is a bandwidth-based estimate for generating text (1024 GB/s); prompt processing and long contexts are slower.

Measured speed on the Instinct MI50 32GB

No community reports for this pair yet.

Report your own speed →

Run GLiDE-27B on the Instinct MI50 32GB

Using Q4_K_M, the best fit for Instinct MI50 32GB.

llama.cpp · OpenAI-compatible server with a web UIdocs
llama-server -hf seamon67/GLiDE-27B-GGUF:Q4_K_M

Needs a Vulkan build of llama.cpp. Downloads the GGUF on first run.

llama.cpp · Docker with Vulkandocs
docker run -d --name llama --device /dev/dri -p 8080:8080 \
  -v ~/.cache/llama.cpp:/root/.cache/llama.cpp \
  ghcr.io/ggml-org/llama.cpp:server-vulkan \
  -hf seamon67/GLiDE-27B-GGUF:Q4_K_M --host 0.0.0.0 --port 8080 -ngl 99

Server and web UI on port 8080.

Transformers · Ask about an imagedocs
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="fastino/GLiDE-27B", dtype="auto", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
            {"type": "text", "text": "Describe this image."},
        ],
    }
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])

Vision models that fit the Instinct MI50 32GB comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
25.8BA4BQ8_0≈29 GBFits in memory~133 tok/sCommercial OK
27.8BQ6_K≈25 GBFits in memory~25 tok/sCommercial OK
27.8BQ6_K≈25 GBFits in memory~25 tok/sCommercial OK
31.3BQ5_K_M≈29 GBFits in memory~25 tok/sCommercial OK
27.4BQ6_K≈26 GBFits in memory~25 tok/sConditions
27.8BQ6_K≈24 GBFits in memory~25 tok/sCommercial OK

Other GPUs that run GLiDE-27B

Frequently asked questions

Can the Instinct MI50 32GB run GLiDE-27B?
Yes. The Instinct MI50 32GB runs GLiDE-27B fully in VRAM at Q8_0 (about 30 GB of 32 GB), at roughly 20 tokens/s.
How fast does GLiDE-27B run on the Instinct MI50 32GB?
An estimate from memory bandwidth gives about 20 tokens/s at Q8_0.
Which quantisation of GLiDE-27B should I use on the Instinct MI50 32GB?
Q8_0: the largest one that fits. Expected quality: lossless compared with the full-precision weights.

Memory and speed are estimates from the model's size and the card's memory bandwidth; real numbers depend on the runtime, drivers and context length. Community figures are medians of user reports.