1× Strix Halo
85.1tok/sPeak
1 request, synthetic prompt Source
Intelligence8with thinkingno thinking 7
Local AI · Recipe · 2× RTX 3090
Qwen3 BF16 with vLLM on 2× RTX 3090
by alesha-pro
No speed value is documented for this recipe, so this page is not indexed.
Intelligence (original model) · Artificial Analysisno independent value
Engine
- Engine
- vLLM 0.26.0
- Quantization
- BF16 (Modelle), Ridge-Mapper fp32/bf16
- Model family
- Qwen3
- Context
- 32,768
- Parameters
- Qwen3-0.6B → Qwen3-1.7B (BF16, 8 KV heads, head_dim 128, 28 Layer)
- Creator
- alesha-pro
- GitHub stars
- 17
- Repo updated
- Aug 10, 2026
Measurements
Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.
Time to first token
Shorter is better.
What you need
- Hardware
- 2 × NVIDIA RTX 3090
- Weights
- BaseQwen/Qwen3-0.6B Apache-2.0
- Main weightsQwen/Qwen3-1.7B Apache-2.0
- Engine
- vLLM 0.26.0
- Context
- 32,768 tokens
Notes
What matters before you rebuild it.
- Experimental
Sources
Related recipes
2× RTX 3090
199tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligenceno independent value
2× RTX 3090
187tok/sEveryday
1 request, prose prompt, prompt 450 tokens, no speculative decoding Source
Intelligenceno independent value
2× RTX 3090
179tok/sEveryday
1 request, prose prompt, no speculative decoding Source
Intelligence18with thinkingno thinking 15
- License unclear
2× RTX 3090
151tok/sEveryday
1 request, prose prompt, no speculative decoding Source
Intelligenceno independent value
2× RTX 3090
144tok/sEveryday
1 request, prose prompt, no speculative decoding Source
Intelligenceno independent value
Local AI in your company?
In a workshop we work out which models and which hardware fit your tasks, and build the first agent on your infrastructure.