Just about. At Q3_K_M it needs about 15 GB of the 16 GB available, so keep the context short, at roughly 13 tokens/s.
- best quant
- Q3_K_M
- needs
- 15 GB
- usable
- 16 GB
- est. speed
- ~13 tok/s
Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the Arete-27B page →
Every quantisation of Arete-27B on the Tesla T4
| Quant | Memory | Fit | Speed |
|---|---|---|---|
| Q2_K | 13 GB | fits | ~20 tok/s |
| Q3_K_M pick | 15 GB | tight | ~13 tok/s |
| Q4_K_M | 19 GB | offload | ~6.1 tok/s |
| Q5_K_M | 23 GB | offload | ~4 tok/s |
| Q6_K | 26 GB | offload | ~3 tok/s |
| Q8_0 | 31 GB | offload | ~2 tok/s |
| FP16 / BF16 | 55 GB | too big | — |
Speed is a bandwidth-based estimate for generating text (320 GB/s); prompt processing and long contexts are slower.
Measured speed on the Tesla T4
No community reports for this pair yet.
Report your own speed →Run Arete-27B on the Tesla T4
Using Q3_K_M, the best fit for Tesla T4.
ollama run hf.co/bartowski/NousResearch_Arete-27B-GGUF:Q3_K_M llama-server -hf bartowski/NousResearch_Arete-27B-GGUF:Q3_K_MDownloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.
lms get bartowski/NousResearch_Arete-27B-GGUF@Q3_K_MThen load it from the LM Studio app or with `lms load`.
vllm serve NousResearch/Arete-27B vllm serve processreap/Arete-27B-FP8services:
vllm:
image: vllm/vllm-openai:latest
command: NousResearch/Arete-27B
ports:
- "8000:8000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 8000.
sglang serve NousResearch/Arete-27B --host 0.0.0.0 --port 30000services:
sglang:
image: lmsysorg/sglang:latest
command: sglang serve NousResearch/Arete-27B --host 0.0.0.0 --port 30000
ports:
- "30000:30000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 30000.
Chat models that fit the Tesla T4 comfortably
All that fit →| Model | Params | Best quant | Memory | Speed |
|---|---|---|---|---|
| 20.9BA4.2B | Q8_0 | ≈13 GBFits in memory | ~40 tok/s | |
| 14.8B | Q6_K | ≈14 GBFits in memory | ~14 tok/s | |
| 14.7B | Q6_K | ≈14 GBFits in memory | ~15 tok/s | |
| 14.8B | Q6_K | ≈14 GBFits in memory | ~14 tok/s | |
| 8.2B | Q8_0 | ≈10 GBFits in memory | ~20 tok/s | |
| 8B | Q8_0 | ≈9.8 GBFits in memory | ~21 tok/s |
Other GPUs that run Arete-27B
Frequently asked questions
- Can the Tesla T4 run Arete-27B?
- Just about. At Q3_K_M it needs about 15 GB of the 16 GB available, so keep the context short, at roughly 13 tokens/s.
- How fast does Arete-27B run on the Tesla T4?
- An estimate from memory bandwidth gives about 13 tokens/s at Q3_K_M.
- Which quantisation of Arete-27B should I use on the Tesla T4?
- Q3_K_M: the largest one that fits. Expected quality: noticeable loss 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.