Inkling Multimodal Model Crashes on Audio Due to Hardcoded Dimensions
MaziyarPanahi · x · 2026-08-01
Developer @MaziyarPanahi discovered that the audio processing feature in the Inkling-Small model crashes. The root cause is a hardcoded nembd = 6144 in the Inkling PR, while Inkling-Small actually uses 4096 dimensions, triggering an assertion error.
Deriving the dimension from the tensor fixes the issue, while image functionality remains unaffected. The patch is being sent upstream.
Related event: Inkling-Small Shines in Medical Parsing but Faces Audio Bug(2 posts)→
More from Models
- Google Unveils Gemini Robotics 2 for Advanced Multi-Finger Dexterous Control — bousmalis · 2026-08-01
- DeepSeek's Suspected V4-Flash Model Endpoint Surfaces on Hugging Face — victormustar · 2026-08-01
- Elon Musk Announces Grok 4.5: Beats GPT-5.6 in Benchmarks, Launches CLI Coding Agent — elonmusk · 2026-08-01
- OpenAI Slashes Model Costs by 80%: Tech Breakthrough or Price War? — Tight-Grocery9053 · 2026-08-01
- Kimi K3 DSpark Upgrade: 1M Context Without Performance Degradation, 140k Downloads — BanghuaZ · 2026-08-01
- Exploring Claude Opus's Odd Visual Outputs with the "Dario and Amanda" Prompt — chicametipo · 2026-08-01