Grok 4.6 Benchmarks: Leads in Coding but Trails in SWE Tasks
kimmonismus · x · 2026-08-12
Users have praised Grok 4.6 for showing an absolutely insane jump in performance.
According to xAI's benchmarks, Grok 4.6 matches GPT-5.6 Sol on the AA Intelligence Index at 61 and leads it on CursorBench, FrontierCode, and AA-Briefcase.
However, it still trails GPT-5.6 Sol on DeepSWE and Terminal-Bench. xAI noted that the model received a longer supplemental training run, regenerated SFT trajectories, and implemented agentic reinforcement learning.
Related event: xAI Launches Grok 4.6: Top Benchmark Performance at Low Cost(27 posts)→
More from Models
- Qwen 1-bit Quantization Shrinks Model to 397GB, a 91% Reduction — danielhanchen · 2026-08-13
- Users Complain About Claude's Unstoppable Chain-of-Thought Output — GuyHachmon · 2026-08-13
- OpenAI's gpt-live-1 Achieves Near-Perfect Conversational Turn Detection — pbbakkum · 2026-08-13
- xAI Launches Grok 4.6 Across Cursor, API with 2x Token Promo — aman_madaan · 2026-08-13
- xAI Exec Hints at Grok 4.5: Focused on Coding Agents — aman_madaan · 2026-08-13
- Reddit Speculates: Is Grok 4.6 a Fine-tune of Kimi K3? — robertpro01 · 2026-08-13