3× DGX Spark
31.3tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence45with thinking
- Custom kernel required
- License unclear
Local AI · Recipe · 4× DGX Spark
GLM-5.3 EXL3 3.38 bpw with TensorFold on 4× DGX Spark
GLM-5.3 EXL3 3.38 bpw with TensorFold on 4× DGX Spark: 35.7 tok/s according to github.com (dataset as of Sep 29, 2026).
by vcruz305
Engine
- Engine
- TensorFold Fork vcruz305/TensorFold glm53-tp4-spark @757a851 (PR #159, TP=4) + b12x RoCE (b58f34e)
- Quantization
- EXL3 SAGE MixedK 3.38 bpw (319 GB; serviert mit BF16-lm_head), bf16 KV, Draft DFlash2 d7/c0.60
- Model family
- GLM-5.3
- Context
- 262,144
- Parameters
- 78 Layer, 256 routed Experten je MoE-Layer; Gesamtparameter in README nicht genannt
- Creator
- vcruz305
- GitHub stars
- 3
- Repo updated
- Oct 5, 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
What you need
- Hardware
- 4 × NVIDIA DGX Spark (GB10)
- Weights
- Main weightsvcruz305/GLM-5.3-EXL3-3.38bpw other
- Draft modelincoai/GLM-5.3-DFlash2 Non-commercial · CC-BY-NC-ND-4.0
- Basezai-org/GLM-5.3 other
- Engine
- TensorFold Fork vcruz305/TensorFold glm53-tp4-spark @757a851 (PR #159, TP=4) + b12x RoCE (b58f34e)
- Context
- 262,144 tokens
Notes
What matters before you rebuild it.
- Non-commercial (draft)
- Custom kernel required
- Fork
- Non-commercial
Sources
Related recipes
1× Mac
6.14tok/sMixed
1 request, mixed prompt set, context 1k, no speculative decoding Source
Intelligence45with thinkingOriginal model's value. Heavily compressed (~2.1 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Streams from SSD
- License unclear
4× DGX Spark
53.3tok/sPeak
1 request, structured output end-to-end (count-to-300); DCP4 lane 2: NVFP4 KV Source
Intelligence45with thinking
- Custom kernel required
- License unclear
4× DGX Spark
87.7tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
4× DGX Spark
64.7tok/sEveryday
1 request, decode at 16K context (profile table), thinking on, xhigh Source
Intelligenceno independent value
- Custom kernel required
4× DGX Spark
64tok/sEveryday
1 request, decode at 16K context (profile table), thinking on, xhigh Source
Intelligence40with thinking
- 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.