Local AI · Recipe · 2× RTX PRO 6000

GLM-5.3-Flash W4A16 (FP8 experts) with SGLang on 2× RTX PRO 6000

GLM-5.3-Flash W4A16 (FP8 experts) with SGLang on 2× RTX PRO 6000: 165 tok/s according to github.com (dataset as of Nov 3, 2026).

by helixml

165tok/sPeak

1 request, best value reported by source, with EAGLE

Source: github.com

Intelligence (original model) · Artificial Analysis42with thinking

Instructions in the README Repo

Engine: SGLang (Image sha256:899fe8eb…, TP2, EAGLE adaptiv: 5 Steps / 6 Draft-Tokens, modelopt_mixed) hinter ramjet v0.8.0
Quantization: W4A16 NVFP4 (K32) + FP8-Experten (ormandj, weight-only), FP8-KV (e4m3), bf16 Mamba/SSM-State
Model family: GLM-5.3-Flash
Context: 524,288
Creator: helixml
GitHub stars: 9
Repo updated: Oct 26, 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

Peak (decode)165 tok/sREADME „Models with a validated stack“: Decode @ c1, TP2; 388,2 tok/s pro 2-GPU-Replica (c4/max256, Box noch nicht saturiert) Source

What you need

Hardware: 2 × NVIDIA RTX PRO 6000 Blackwell
Weights: Main weightsormandj/GLM-5.3-Flash-W4A16-NVFP4-K32-Experts-FP8-WO MIT
Basezai-org/GLM-5.3-Flash MIT
Engine: SGLang (Image sha256:899fe8eb…, TP2, EAGLE adaptiv: 5 Steps / 6 Draft-Tokens, modelopt_mixed) hinter ramjet v0.8.0
Context: 524,288 tokens

Sources

Related recipes

4× DGX Spark

GLM-5.3-Flash NVFP4/MXFP8 CSF with SparkRing on 4× DGX Spark

by fujitsupolycom

68tok/sEveryday

1 request, generation at 32K context (speed table; measured only as a path of four Sparks inside a ring of eight, not as a ring of four), thinking always on (measured before the profile default high, i.e. template default max), with MTP3 (per recipe) Source

Intelligence42with thinking

Custom kernel required
License unclear

NVFP4/MXFP8 CSF QAD (Local Inference Lab)vLLM (local-inference-lab Fork) via SparkRing (SIRCL-Collectives, switchless RDMA ueber ConnectX-7), Image Release 2026.10.2Weights Repo Updated Nov 2, 2026

Details

4× DGX Spark

GLM-5.3-Flash EXL3 4bpw with TensorFold on 4× DGX Spark

by tonyd2wild

65.4tok/sEveryday

1 request, prose prompt, with MTP + DFlash2 (per recipe) Source

Intelligence42with thinking

Custom kernel required
Non-commercial
License unclear

EXL3 4bpw routed experts (mcg) + q4 dense (at load) + FP8 KVTensorFold v0.6.0 + MiaAI-Lab GLM-5.3-Flash patch series (engine-multi, + tonyd2wild changes)Weights Repo Updated Oct 12, 2026

Details

2× DGX Spark

GLM-5.3-Flash EXL3 with TensorFold on 2× DGX Spark (1M context, 4 streams)

by MiaAI-Lab

60.4tok/sEveryday

Prose decode, 1 request; quality (FP8 KV, 4-bit dense): GSM8K 98.8% / HumanEval 95.7% / HumanEval+ & MBPP+ 86.5% Source

Intelligence42with thinking

≤ 3 bit: quantization may cost quality
Custom kernel required
Non-commercial
License unclear

EXL3/TR3 4bpw routed experts + 4-bit dense, BF16 elsewhere, FP8 KVTensorFold v0.6.0 + Recipe v1.10 (README: 96 Patches; gepinntes Image v0.6.0-e66c657a433d mit 107 Patches: DFlash2+Copy-Drafts, 4-bit dense, FP8-KV, RoCE-One-Shot-All-Gather, Shared KV-Pool, optionaler Spill-Tier, Vision, Tool-Calling, Anthropic /v1/messages; PARALLEL=8 opt-in)Weights Repo Updated Oct 25, 2026

Details

2× DGX Spark

GLM-5.3-Flash EXL3 4bpw with TensorFold on 2× DGX Spark

by tonyd2wild

44tok/sEveryday

1 request, prose prompt, with MTP + DFlash2 (per recipe) Source

Intelligence42with thinking

Custom kernel required
Non-commercial
License unclear

EXL3 4bpw routed experts (mcg) + q4 dense (at load) + FP8 KVTensorFold v0.6.0 + MiaAI-Lab GLM-5.3-Flash patch series (engine-multi, + tonyd2wild changes)Weights Repo Updated Oct 12, 2026

Details

4× DGX Spark

GLM-5.3-Flash 320B-A18B NVFP4 with vLLM on 4× DGX Spark

by tonyd2wild

40.8tok/sEveryday

1 request, prose prompt, with DFlash2 (per recipe) Source

Intelligence42with thinking

Custom kernel required
Non-commercial

NVFP4 weights + FP8 KV (Lane B: NVFP4 KV)vLLM 0.28.0Weights Repo Updated Oct 10, 2026

Details

3× DGX Spark

GLM-5.3-Flash 320B EXL3 4 bpw with vLLM on 3× DGX Spark

by MiaAI-Lab

40.1tok/sEveryday

1 request, prose prompt, with DFlash2, TR3 weights (per recipe) Source

Intelligence42with thinking

≤ 3 bit: quantization may cost quality
Custom kernel required
Non-commercial
License unclear

EXL3/TR3 4 bpw (routed experts), FP8 KV (fp8_ds_mla)vLLMWeights Repo Updated Oct 18, 2026

Details

← Back to overview

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.

See the AI workshop Back to overview

Engine

Measurements

Decode · 1 request

What you need

Sources

Related recipes

GLM-5.3-Flash NVFP4/MXFP8 CSF with SparkRing on 4× DGX Spark

GLM-5.3-Flash EXL3 4bpw with TensorFold on 4× DGX Spark

GLM-5.3-Flash EXL3 with TensorFold on 2× DGX Spark (1M context, 4 streams)

GLM-5.3-Flash EXL3 4bpw with TensorFold on 2× DGX Spark

GLM-5.3-Flash 320B-A18B NVFP4 with vLLM on 4× DGX Spark

GLM-5.3-Flash 320B EXL3 4 bpw with vLLM on 3× DGX Spark

Local AI in your company?