Org card: add GLM-5.2 W4A16-MTP as lead release

README.md

@@ -32,6 +32,32 @@ Web · [cql.ca](https://cql.ca)
 
 ---
 
+## Latest release — GLM-5.2 W4A16-MTP
+
+A 4-bit weight quantization of [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2) (744B-parameter MoE) that keeps the multi-token-prediction (MTP) draft head in BF16. It **matches the FP8 release on quality, fits on four H200s instead of eight** (~1.49 TB BF16 → ~405 GB), and is the **fastest of the popular 4-bit GLM-5.2 quants in the interactive serving regime**. MIT-licensed.
+
+→ [**canada-quant/GLM-5.2-W4A16-MTP**](https://huggingface.co/canada-quant/GLM-5.2-W4A16-MTP)
+
+**Recipe** — routed-expert weights to INT4 (group-size 128, GPTQ, via llm-compressor); attention, dense prefix layers, shared experts, router, embeddings, and LM head left in BF16. The MTP draft head is re-injected at BF16 after quantization, so speculative decoding survives end-to-end — a lossless speedup that changes latency, not answers.
+
+**Quality** — within run-to-run noise of `zai-org/GLM-5.2-FP8` on the same harness (8×H200):
+
+| Task | W4A16+MTP | FP8 |
+|---|---|---|
+| GSM8K (strict) | 0.960 | 0.955 |
+| IFEval (prompt / inst, strict) | 0.909 / 0.911 | 0.891 / 0.903 |
+| MATH-500 | 0.954 | 0.958 |
+| RULER @ 32K / 64K | 0.832 / 0.841 | 0.831 / 0.813 |
+| SWE-bench Verified | 82.0% | 82.2% |
+
+**Speed** — 132 output tok/s at concurrency 1, **+69% over the next-fastest 4-bit GLM-5.2 quant**; +48% vs FP8 at c=1 and +32% at c=8, where MTP helps most. At full saturation the no-MTP quants pull ~13–15% ahead — an honest trade-off in the high-throughput regime.
+
+**Serving** — 4×H200 covers up to ~128K context; the full 1M-token context needs 8×H200. Validated on Hopper (H200); Blackwell serving needs additional kernel flags. Built on [GLM-5.2](https://huggingface.co/zai-org/GLM-5.2), quantized with [llm-compressor](https://github.com/vllm-project/llm-compressor), served with [vLLM](https://github.com/vllm-project/vllm). Full recipe, evaluation methodology, and engineering log in the repo.
+
+Writeup: [Running GLM-5.2 on half the GPUs: a W4A16 + MTP quantization](https://cql.ca/news/glm-5-2-w4a16-mtp.html).
+
+---
+
 ## Open releases — DeepSeek-V4 quantization family
 
 Four artifacts in the same lineage. One base model in two sizes (V4-Flash, V4-Pro); two routed-expert formats (W4A16, NVFP4); Multi-Token Prediction (MTP) draft head retained on three of four. Attention is FP8 block 128×128 across all four.