DeepSeek-V3.2vsGrok 4.6
DeepSeek-V3.2 | Grok 4.6 | |
|---|---|---|
| Benchmarks | ||
Nonsense detection BullshitBench v2Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 13% | — |
Agentic coding CursorBench v3.2Cursor's own test of harder, real-world coding tasks inside a code editor, on the refreshed v3.2 task set. Scores aren't comparable with v3.1. Higher is better. | — | 69.9% |
Agentic coding DeepSWE 1.1Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. | — | 65.9% |
Agentic coding FrontierCode v1.1 (Extended) · extended splitfrontier-difficulty agentic coding tasks (v1.1, extended split) | — | 61.3% |
Expert software engineering APEX-SWEexpert-level software-engineering tasks (AI Productivity Index) | — | 56.4% |
Agentic terminal coding Terminal-Bench 3.0command-line task completion (v3.0, much harder task set) | — | 26% |
Expert agentic work APEX-Agentsexpert-level agentic work tasks (AI Productivity Index) | — | 57.5% |
Agentic legal work Harvey's Legal Agent BenchmarkHarvey's test of whether an AI agent can complete real legal work — drafting and reviewing documents, working with spreadsheets and presentations, and navigating files the way a lawyer's assistant would. Higher is better. | — | 15.8% |
Overall intelligence AA Intelligence IndexArtificial Analysis composite intelligence index across evals | — | 61 |
Knowledge work GDPval-AA v2economically valuable knowledge work (v2, re-based Elo) | — | 1753 |
Knowledge work AA-BriefcaseArtificial Analysis agentic office-work eval (Elo) | — | 1577 |
Community preference (code) Arena Elo (Code)Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better. | 1361 | — |
| Overview | ||
| Company | DeepSeek | SpaceXAI |
| Release date | Dec 1 2025 | Aug 12 2026 |
| Access | Open Weight | Proprietary |
Which is better: DeepSeek-V3.2 or Grok 4.6?
DeepSeek-V3.2 and Grok 4.6 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. DeepSeek-V3.2 shipped 254 days before Grok 4.6, so benchmark comparisons should account for the intervening progress.
DeepSeek-V3.2 is open weight, while Grok 4.6 is proprietary.
Direct benchmark comparisons are unavailable — DeepSeek-V3.2 and Grok 4.6 don't publish scores on any of the same benchmarks.
Frequently asked questions
DeepSeek-V3.2 was released by DeepSeek on Dec 1 2025.
Grok 4.6 was released by SpaceXAI on Aug 12 2026.
DeepSeek-V3.2 is an open weight model released by DeepSeek. Grok 4.6 is a proprietary model released by SpaceXAI.
Other comparisons
DeepSeek-V3.2vsClaude Opus 5Grok 4.6vsClaude Opus 5DeepSeek-V3.2vsGPT-5.6-CyberGrok 4.6vsGPT-5.6-CyberDeepSeek-V3.2vsGemini 3.7 FlashGrok 4.6vsGemini 3.7 FlashDeepSeek-V3.2vsMuse GlimmerGrok 4.6vsMuse GlimmerDeepSeek-V3.2vsMistral Medium 3.5Grok 4.6vsMistral Medium 3.5DeepSeek-V3.2vsKimi K3Grok 4.6vsKimi K3