Open-source 4B Smaug-Audio-Decisions tops audio decision leaderboard, beating Gemini APIs
bindureddy · x · 2026-11-04
Bindu Reddy announced the open-source Smaug-Audio-Decisions, a 4B decision model that uniquely accepts audio, image and text inputs. Benchmark chart shows it leading its class with audio 60.1, images 57.9, text 56.6 — its audio score beating even Gemini 4 Argon (59.4) and Gemini 3.8 Flash (50.5) API reference runs. Available now on Hugging Face for voice agents.
More from Models
- Model Vendors Post-Train for Tool Search, Squeezing Out MCP-Layer Workarounds — pvncher · 2026-11-04
- Claude Opus 5.5 blocks viewing Qwen's reasoning, safeguard error says sorry not allowed — andpoul · 2026-11-04
- Gemini API deprecations: 2.5 Pro auto-redirects to 3.8 Flash on Dec 1 — patloeber · 2026-11-04
- Tencent's Hy4 coming to LMArena after preview hit #2 among open models in Agent Arena — arena · 2026-11-04
- Mistral Large 4 debuts with 1.05T open weights and 1M context on Open Math OS — MistralAI · 2026-11-04
- Hugging Face invites Datalab's OmniParseBench to become a native hub benchmark — NielsRogge · 2026-11-04