diff --git a/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml b/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml
index 030ae618f..ab0f4d024 100644
--- a/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml
+++ b/recipes/glm-5-3-flash-exlr8-exl3-experts-on-dgx-spark-tp1-4/recipe.yaml
@@ -7,7 +7,7 @@ engine:
   git_url: 'https://github.com/kindlingai/vllm-exlr8'
   pinned_commit: a2e163cb05
   install_note: 'Source build: docker build -f image/Dockerfile . The overlay sets --quantization exlr8 --load-format exlr8 over the NVFP4 Flash base image.'
-summary: "GLM-5.3-Flash exlr8: MoE layers are EXL3-trellis units decoded by kindlingai's custom exlr8 MoE kernel; dense linears stay FP8/NVFP4. One assembled release carries three expert widths (k2/k3/k4 = 2/3/4 bpw), chosen at boot via VLLM_EXLR8_WIDTHS. Stack: kindlingai/vllm-exlr8 over the NVFP4 Flash base recipe, with DFlash2 speculation, adaptive-k, RecoverSSM, sparse MLA and mentat unchanged. Reference topology TP=4 on four GB10 boxes; TP=1 serves the whole model (~87 GiB) on one 128 GB DGX Spark / ASUS GX10 using tp1.yaml plus a distinct MENTAT_GROUP. TP=2 and TP=3 also measured. Author-measured 2026-10-19, clocks TP-locked at 2200 MHz: decode code 128.0 tok/s and prose 72.5 at TP=4; 53.0 / 29.5 at TP=1; prefill 4,393 tok/s at 8K and 4,190 at 32K. Trade vs the shipping NVFP4 build: decode +3-8%, NLL equal or better, KV pool ~+15%, prefill 7.7% lower at 32k/128k."
+summary: 'GLM-5.3-Flash exlr8: MoE layers are EXL3-trellis units decoded by the custom exlr8 MoE kernel; dense linears stay FP8/NVFP4. One release carries three expert widths (k2/k3/k4 = 2/3/4 bpw), chosen at boot via VLLM_EXLR8_WIDTHS. Reference topology TP=4 on four GB10 boxes; TP=1 serves the whole model (~87 GiB) on one 128 GB DGX Spark / ASUS GX10 via tp1.yaml plus a distinct MENTAT_GROUP. TP=2 and TP=3 also measured. 2026-10-19, clocks TP-locked at 2200 MHz: decode code 128.0 tok/s, prose 72.5 at TP=4; 53.0 / 29.5 at TP=1; prefill 4,393 tok/s at 8K, 4,190 at 32K. Trade vs the shipping NVFP4 build: decode +3-8%, NLL equal or better, KV pool ~+15%, prefill 7.7% lower at Second 1x-Spark GLM-5.3-Flash lane: not the VQ Q4_K GGUF lane (/r/glm-5-3-flash-vq-q4-k-1x-dgx-spark-1m-ctx-jlfn), and not a fork of the 2x ATLAS build (/r/glm-5-3-flash-nvfp4-on-2x-dgx-spark-atlas-rc4-2-tp2-dflash2-1m) whose NVFP4 Flash base this stack overlays.'
 license: MIT
 model:
   name: GLM-5.3-Flash-EXL3
@@ -142,11 +142,11 @@ artifacts:
 provenance:
   credit:
     -
-      name: 'kindlingai (exlr8 recipe and vllm-exlr8 fork)'
-      url: 'https://github.com/kindlingai'
+      name: kindlingai
+      url: 'https://github.com/kindlingai/glm-5.3-flash-exlr8-gx10'
     -
-      name: 'mmastrac (NVFP4 Flash 4x-GX10 base recipe)'
-      url: 'https://github.com/mmastrac'
+      name: nvidia
+      url: 'https://huggingface.co/nvidia/GLM-5.3-Flash-NVFP4'
   source_urls:
     - 'https://github.com/kindlingai/glm-5.3-flash-exlr8-gx10'
     - 'https://github.com/mmastrac/glm-5.3-flash-4x-gx10'