Skip to main content

Grok 4.6 vs Grok 4.7

SpecGrok 4.6Grok 4.7
Respan Index53.4 (#32)53.0 (#34)
Blended price per 1M$3$3
Context window500K500K
Input / output per 1M$2 / $6$2 / $6
Max output--
Inputtext, visiontext, vision
Released2026-08-122026-09-21

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryGrok 4.6Grok 4.7
Coding34.0 Grok 4.642.6 Grok 4.7
Agents65.8 Grok 4.670.4 Grok 4.7
Reasoning59.4 Grok 4.642.7 Grok 4.7
Knowledge & language51.1 Grok 4.655.5 Grok 4.7
Human preference56.5 Grok 4.656.9 Grok 4.7

How the Respan Index works

Benchmarks both report

Grok 4.7 leads on 15 of the 26 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkGrok 4.6Grok 4.7
LMArena Elo14571443
LMArena Vision (LMArena)12641250
LMArena WebDev (LMArena)16181638
LMArena Agent (LMArena)-0.0060.0148
MirrorCode (Epoch AI)41.1%54.4%
Chess Puzzles (Epoch AI)33%41%
APEX-Agents (Mercor)65.3%70%
FrontierMath Tier 4 v2 (Epoch AI)31.7%17.1%
FrontierMath Tiers 1-3 v2 (Epoch AI)66%53%
FrontierSWE V2 (Proximal Labs)25.3%29.5%
Furniture Assembly (Epoch AI)40%20.8%
GPQA Diamond94%92.7%
LiveBench Agentic Coding (LiveBench)57%54%
LiveBench Coding (LiveBench)76.8%77.2%
LiveBench Data Analysis (LiveBench)73.9%76.9%
LiveBench Instruction Following (LiveBench)71.9%75.3%
LiveBench Language (LiveBench)83.7%80.1%
LiveBench Mathematics (LiveBench)92.6%95.7%
LiveBench Reasoning (LiveBench)90.5%82.7%
Mystery Game Puzzles (Epoch AI)34%29%
OTIS Mock AIME 2024-2025 (Epoch AI)99.2%98.1%
SAGE (Vals AI)28.9%40.8%
SimpleQA Verified49.3%56%
Terminal-Bench 4.0 (Vals AI)17.2%28.8%
Vending-Bench 2 (Andon Labs)9047.0310536.83
DeepSWE v1.165.9%71%

FAQ

Is Grok 4.6 better than Grok 4.7?

Grok 4.6 scores 53.4 on the Respan Index (#32), ahead of Grok 4.7 at 53.0 (#34). Grok 4.7 leads on 15 of the 26 benchmarks both report.

Which is cheaper, Grok 4.6 or Grok 4.7?

Grok 4.6 lists at $2 input and $6 output per 1M tokens; Grok 4.7 at $2 and $6. Both cost about the same per token at list price.

Which has the larger context window?

Both accept 500K tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.