4× DGX Spark
53.3tok/sPeak
1 request, JSON prompt Source
Intelligence45with thinking
- Custom kernel required
- License unclear
Local AI · Recipe · 1× Mac
GLM-5.3 UD-IQ2_XXS with DwarfStar on 1× Mac
by antirez
Intelligence (original model) · Artificial Analysis45with thinkingOriginal model's value. Heavily compressed (~2.1 bit), intelligence is likely much lower.
≤ 3 bit: quantization may cost quality
Engine
- Engine
- DwarfStar (ds4) main@0aaea5a (2026-09-20)
- Quantization
- UD-IQ2_XXS routed, blk78 Q2K, 196.58 GiB
- Model family
- GLM-5.3
- Context
- —
- Creator
- antirez
- GitHub stars
- 22,692
- Repo updated
- Sep 25, 2026
Measurements
Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.
No sourced measurements for this recipe.
What you need
- Hardware
- 1 × Mac / Apple Silicon
- Weights
- Main weightsantirez/glm-5.3-gguf License unclear
- Engine
- DwarfStar (ds4) main@0aaea5a (2026-09-20)
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.
- Streams from SSD
- License unclear
Sources
Related recipes
1× Mac
25.1tok/sEveryday
1 request, prose prompt, short prompt, with MTP (DSpark built-in) (per recipe) Source
Intelligence40with thinkingno thinking 25
- ≤ 3 bit: quantization may cost quality
- No license on HF
- License unclear
1× Mac
123tok/sPeak
1 request, best value reported by source Source
Intelligence9with thinking
1× Mac
102tok/sPeak
1 request, best value reported by source Source
Intelligence8with thinkingno thinking 7
1× Mac
48.7tok/sPeak
1 request, best value reported by source, context 32k Source
Intelligence40with thinkingOriginal model's value. Heavily compressed (~2.1 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- License unclear
1× Mac
48.4tok/sPeak
1 request, best value reported by source, context 32k Source
Intelligence40with thinking
- License unclear
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.