Can the RTX 3090 run Apriel-Voice-27B-preview?
Apriel-Voice-27B-preview·27.2B·GeForce RTX 3090·24 GB VRAM
Yes. The RTX 3090 runs Apriel-Voice-27B-preview fully in VRAM at Q5_K_M (about 21 GB of 24 GB), at roughly 27 tokens/s.
- best quant
- Q5_K_M
- needs
- 21 GB
- usable
- 24 GB
- est. speed
- ~27 tok/s
Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the Apriel-Voice-27B-preview page →
Every quantisation of Apriel-Voice-27B-preview on the RTX 3090
| Quant | Memory | Fit | Speed |
|---|---|---|---|
| Q2_K | 11 GB | fits | ~58 tok/s |
| Q3_K_M | 15 GB | fits | ~39 tok/s |
| Q4_K_M | 19 GB | fits | ~31 tok/s |
| Q5_K_M pick | 21 GB | fits | ~27 tok/s |
| Q6_K | 24 GB | offload | ~19 tok/s |
| Q8_0 | 31 GB | offload | ~4.3 tok/s |
| FP16 / BF16 | 56 GB | too big | — |
Speed is a bandwidth-based estimate for generating text (936 GB/s); prompt processing and long contexts are slower.
Measured speed on the RTX 3090
No community reports for this pair yet.
Report your own speed →Omni models that fit the RTX 3090 comfortably
All that fit →| Model | Params | Best quant | Memory | Speed |
|---|---|---|---|---|
| 35.3BA3B | Q4_K_M | ≈19 GBFits in memory | ~283 tok/s | |
| 24.3B | Q6_K | ≈21 GBFits in memory | ~26 tok/s | |
| 31.7BA3B | Q4_K_M | ≈19 GBFits in memory | ~283 tok/s | |
| 33BA3B | Q4_K_M | ≈20 GBFits in memory | ~283 tok/s | |
| 31.7BA3B | Q4_K_M | ≈19 GBFits in memory | ~283 tok/s | |
| 17.5B | Q8_0 | ≈20 GBFits in memory | ~28 tok/s |
Other GPUs that run Apriel-Voice-27B-preview
Frequently asked questions
- Can the RTX 3090 run Apriel-Voice-27B-preview?
- Yes. The RTX 3090 runs Apriel-Voice-27B-preview fully in VRAM at Q5_K_M (about 21 GB of 24 GB), at roughly 27 tokens/s.
- How fast does Apriel-Voice-27B-preview run on the RTX 3090?
- An estimate from memory bandwidth gives about 27 tokens/s at Q5_K_M.
- Which quantisation of Apriel-Voice-27B-preview should I use on the RTX 3090?
- Q5_K_M: the largest one that fits. Expected quality: near-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.