4× DGX Spark
87.7tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
Local AI · Recipe · 1× Mac
DeepSeek-V4.1-Flash 763B with MLX on 1× Mac
DeepSeek-V4.1-Flash 763B with MLX on 1× Mac: 25.1 tok/s according to github.com (dataset as of Nov 3, 2026).
by drowzeys
25.1tok/sEveryday
1 request, prose prompt, short prompt, with MTP (DSpark built-in) (per recipe)
Source: github.comIntelligence (original model) · Artificial Analysis40with thinkingno thinking 25
≤ 3 bit: quantization may cost quality
Engine
- Engine
- oMLX 0.7.0.dev2 @395ec2fd, MLX 0.32.2
- Quantization
- oQ3e (affine 3-bit) mit 27/40 MoE-Layern 2-bit gs64
- Model family
- DeepSeek-V4.1-Flash
- Context
- 404,805
- Parameters
- 763B MoE
- Creator
- drowzeys
- GitHub stars
- 6
- Repo updated
- Sep 16, 2026
Measurements
Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.
Decode · 1 request
Prefill by context
Time to first token
Shorter is better.
What you need
- Hardware
- 1 × Mac / Apple Silicon
- Weights
- Main weightsdrowzeys/keys-Mac-oMLX-0.7.0.dev2-DeepSeek-V4.1-Flash-oQ3e-2b27-CED-MTP License unclear
- BaseJundot/DeepSeek-V4.1-Flash-oQ3e-mtp License unclear
- Basedeepseek-ai/DeepSeek-V4.1-Flash MIT
- Engine
- oMLX 0.7.0.dev2 @395ec2fd, MLX 0.32.2
- Context
- 404,805 tokens
Notes
What matters before you rebuild it.
- ≤ 3 bit: quantization may cost qualityAt 3 bit and below the model may answer noticeably worse than the original. The intelligence number refers to the original.
- No license on HF
- License unclear
Sources
Related recipes
8× DGX Spark
74tok/sEveryday
1 request, generation at 32K context (speed table, ring of eight), thinking on, high, with DSpark (per recipe); experimental profile Source
Intelligence40with thinkingno thinking 25
- Custom kernel required
- Experimental
4× RTX PRO 6000
73.1tok/sEveryday
1 request, prose prompt, context 32788, with DSPARK Source
Intelligence40with thinkingno thinking 25
4× DGX Spark
71tok/sEveryday
1 request, decode at 32K context (speed table, ring of four), thinking on, high Source
Intelligence40with thinkingno thinking 25
- Custom kernel required
4× DGX Spark
59.1tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
3× DGX Spark
51tok/sEveryday
1 request, prose prompt, with MTP/EAGLE (spekulativ, ~59 ms/step) (per recipe) Source
Intelligence40with thinkingno thinking 25
- Custom kernel required
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.