InferenceBench update: Gemini 4 Carbon takes #3, breaking Anthropic's leaderboard sweep
maksym_andr · x · 2026-11-17
After InferenceBench added Gemini 4 Argon, Carbon, and Hydrogen plus Gemini 3.8 Flash, Gemini 4 Carbon (High, via OpenCode) ranks #3 with a 10.48x aggregate score and Gemini 4 Hydrogen #8 at 8.93x, breaking into a top-10 previously dominated by Anthropic's Claude Opus 5.5 (Max, 12.08x) and Claude Sonnet 5.5. The benchmark scores prefill, decode, throughput and all-in-one inference performance.
Related event: Gemini 4 Carbon Breaks Into InferenceBench Top Three(2 posts)→
More from Models
- Diffusion images lack controllability: researchers argue for painter-style, tool-call-based generation — josephdviviano · 2026-11-17
- User Has Spent $2,513 on ChatGPT Since 2023 but Eyes a Switch to Claude — MoreLunch1547 · 2026-11-17
- Claude Fable 5.2 writes 43% shorter replies and cuts em dash usage by 99% — ArtificialAnlys · 2026-11-17
- ChatGPT hallucinates an entire nonexistent client email while sorting the user's inbox — jdjohnson · 2026-11-17
- NVIDIA shares guide to run GLM-5.3 NVFP4 locally on two DGX Stations with vLLM — vllm_project · 2026-11-17
- OpenAI quietly adds a remote switch controlling how Codex runs agent teams — ryanmerket · 2026-11-17