[docs]: catalog the two-Spark FastH3 recipe on the MiniMax H3 family page

The pair YAML was already in this PR; the cookbook gallery only lists recipes
that declare a family and hardware evidence.

Author: aryan5v

docs/assets/cookbook-recipes.json

@@ -638,11 +638,26 @@
     },
     {
       "id": "fasth3-spark-pair",
-      "task": "Text to video",
+      "family": "minimax_h3",
+      "stage": "inference",
+      "task": "Few-step text to video (with audio)",
       "label": "FastH3 on two DGX Sparks (sequence parallel)",
+      "summary": "Run one FastH3 clip across two GB10s with Ray sequence parallel over QSFP RoCE. Sequential load and lazy module load stay on because SP replicates the DiT on each node.",
       "model": "FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree",
       "source": "examples/inference/basic/basic_fasth3_spark_pair.yaml",
-      "command": "source examples/inference/optimizations/spark_pair_env.sh && FASTVIDEO_VSA_SM100A=0 FASTVIDEO_FA4=0 FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN_H3 FASTVIDEO_VAE_PARALLEL_DECODE=1 fastvideo generate --config examples/inference/basic/basic_fasth3_spark_pair.yaml"
+      "command": "source examples/inference/optimizations/spark_pair_env.sh && FASTVIDEO_VSA_SM100A=0 FASTVIDEO_FA4=0 FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN_H3 FASTVIDEO_VAE_PARALLEL_DECODE=1 fastvideo generate --config examples/inference/basic/basic_fasth3_spark_pair.yaml",
+      "gpu_types": ["NVIDIA"],
+      "hardware": {
+        "platform": "cuda",
+        "gpu_count": 2,
+        "accelerator": "NVIDIA GB10 (DGX Spark pair)",
+        "evidence": "validated",
+        "evidence_url": "https://github.com/hao-ai-lab/FastVideo/pull/1803"
+      },
+      "evidence": "Verified",
+      "expected_artifact": "MP4 under outputs/fasth3_spark_pair/",
+      "modes": ["T2VA", "2-Spark SP"],
+      "limitations": ["Requires a two-node Ray cluster on the QSFP interconnect. See docs/getting_started/installation/spark_pair.md."]
     },
     {
       "id": "matrix-game-2",