xAI
Grok 4.1
Latest xAI. Heavy RL + tool-use training. 56.8% SWE-Bench Pro.
9 reviews
Community Ratings
Used this model? Share your experience.
Write a ReviewReviews (9)
Frontier tool-calling from heavy RL training. Half the hallucination rate of Grok 4 Fast.
Frontier tool-calling from heavy RL training. Half the hallucination rate of Grok 4 Fast.
Frontier tool-calling from heavy RL training. Half the hallucination rate of Grok 4 Fast.
1722 Elo on Creative Writing v3 — 600 points above xAI's previous best. Only GPT-5.1 is higher.
1722 Elo on Creative Writing v3 — 600 points above xAI's previous best. Only GPT-5.1 is higher.
1722 Elo on Creative Writing v3 — 600 points above xAI's previous best. Only GPT-5.1 is higher.
#1 LMArena with 1483 Elo. Coding is decent but official benchmarks skip SWE-bench suspiciously.
#1 LMArena with 1483 Elo. Coding is decent but official benchmarks skip SWE-bench suspiciously.