Rahul Tuli
AI & ML interests
None yet
Recent Activity
updated a model about 12 hours ago
RelaxingSnorlax/Qwen3.8-27B-speculator.dspark-hs15 updated a model about 12 hours ago
RelaxingSnorlax/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B.dspark-hs15 updated a model about 12 hours ago
RelaxingSnorlax/Qwen3.8-27B-speculator.dspark-hs10Organizations
GLM5.3-Flash
FP8 quantized DSpark speculators for zai-org/GLM-5.3-Flash.
Qwen3.8-27B DSpark Drafters — FP8_DYNAMIC (dummy)
Dummy FP8_DYNAMIC DSpark drafters (5/15/30 layers) cycled from the 5-layer base. For size/latency experiments only.
Qwen3.8-27B DSpark Drafters — BF16 (dummy layer sweep)
Dummy BF16 DSpark drafters (5/15/30 layers) cycled from the 5-layer base. For size/latency experiments only.
GLM5.3-Flash Layer Sweep
Dummy 10/15/30-layer expansions of the GLM-5.3-Flash DSpark drafter (BF16, FP8_DYNAMIC, FP8_BLOCK); cycled layers, not trained.
-
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-BF16-10L
Text Generation • 4B • Updated • 163 -
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-BF16-15L
Text Generation • 5B • Updated • 166 -
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-BF16-30L
Text Generation • 8B • Updated • 167 -
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-FP8-dynamic-10L
Text Generation • 4B • Updated • 70
nemotron-drafters
Nemotron 3.5 DSpark drafters in BF16, FP8_BLOCK, and FP8_DYNAMIC at 5, 15, and 30 layers; expanded variants are dummy layer repeats.
Qwen3.8-27B DSpark Drafters — FP8_BLOCK (dummy layer sweep)
Dummy FP8_BLOCK DSpark drafters (5/15/30 layers) cycled from the 5-layer base. For size/latency experiments only.
aux-ablation
GLM5.3-Flash Layer Sweep
Dummy 10/15/30-layer expansions of the GLM-5.3-Flash DSpark drafter (BF16, FP8_DYNAMIC, FP8_BLOCK); cycled layers, not trained.
-
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-BF16-10L
Text Generation • 4B • Updated • 163 -
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-BF16-15L
Text Generation • 5B • Updated • 166 -
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-BF16-30L
Text Generation • 8B • Updated • 167 -
RelaxingSnorlax/GLM-5.3-Flash-speculator.dspark-preview-FP8-dynamic-10L
Text Generation • 4B • Updated • 70
GLM5.3-Flash
FP8 quantized DSpark speculators for zai-org/GLM-5.3-Flash.
nemotron-drafters
Nemotron 3.5 DSpark drafters in BF16, FP8_BLOCK, and FP8_DYNAMIC at 5, 15, and 30 layers; expanded variants are dummy layer repeats.
Qwen3.8-27B DSpark Drafters — FP8_DYNAMIC (dummy)
Dummy FP8_DYNAMIC DSpark drafters (5/15/30 layers) cycled from the 5-layer base. For size/latency experiments only.
Qwen3.8-27B DSpark Drafters — FP8_BLOCK (dummy layer sweep)
Dummy FP8_BLOCK DSpark drafters (5/15/30 layers) cycled from the 5-layer base. For size/latency experiments only.
Qwen3.8-27B DSpark Drafters — BF16 (dummy layer sweep)
Dummy BF16 DSpark drafters (5/15/30 layers) cycled from the 5-layer base. For size/latency experiments only.