Model reference · open weights
Qwen3 is an open-weight language model from PrimeIntellect. Qwen3-1.7B (BF16) weighs 4.1 GB; the smallest configuration that runs it is RTX 3060 12 GB.
What it is
| Released by | PrimeIntellect |
|---|---|
| Released | 2025-08-12 |
| Parameters | 752M |
| VRAM | 4.1 GB for the weights |
What it runs on
| Card | Requests at once | Context max | Memory | |
|---|---|---|---|---|
| 8K each | 32K each | |||
| RTX 3060 12 GB | 7 | 1 | all 40K | 11.6 GB |
| RTX 4060 Ti 16 GB | 11 | 2 | all 40K | 15.4 GB |
| RTX 3090 24 GB | 19 | 4 | all 40K | 23.4 GB |
| RTX 4090 24 GB | 19 | 4 | all 40K | 23.4 GB |
| RTX 5090 32 GB | 27 | 6 | all 40K | 31.0 GB |
| L40S 48 GB | 41 | 10 | all 40K | 44.0 GB |
| A100 80 GB | 78 | 19 | all 40K | 78.2 GB |
| H100 80 GB | 73 | 18 | all 40K | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | 90 | 22 | all 40K | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | 104 | 26 | all 40K | 107 GB |
| H200 141 GB | 137 | 34 | all 40K | 138 GB |
| B200 180 GB | 177 | 44 | all 40K | 176 GB |
| 2× RTX 3060 12 GB tensor parallel | 18 | 4 | all 40K | 11.6 GB a card |
| Requests at once | 8K tokens each | 32K tokens each |
|---|---|---|
| 1 | 5.7 GB | 8.5 GB |
| 5 | 9.4 GB | 23.5 GB |
| 8 | 12.2 GB | 34.8 GB |
| 16 | 19.8 GB | 64.9 GB |
| 32 | 34.8 GB | 125 GB |
| 64 | 64.9 GB | 245 GB |
One card, with vLLM's small-card settings.
Builds
From the model card
This is a clone of Qwen/Qwen3-0.6B with a multi-turn, tool-call compatible chat template.
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.