text model comparison
Grok 4.20 vs Grok 4.6
Grok 4.20 and Grok 4.6 are both current text models, from xAI. Grok 4.20 is 38% cheaper per 1M input tokens than Grok 4.6, $1.25 vs $2. On Arena Text, Grok 4.20 scores 1472 Elo and Grok 4.6 1454 Elo (Arena, measured October 8, 2026). Grok 4.20 has the larger context window, 1,000,000 vs 500,000 tokens.
Data verified October 12, 2026 · Dataset 2026-10-12
Side by side
Standard-tier prices in effect today. The better value in each row is highlighted.
| Fact | Grok 4.20 | Grok 4.6 |
|---|---|---|
| Price per 1M input tokens | $1.25 (better) | $2 |
| Price per 1M output tokens | $2.50 (better) | $6 |
| Cost of 10M input and 2M output tokens | $17.50 (better) | $32 |
| Arena Text | 1472 Elo (#22 of 39) (better) | 1454 Elo (#28 of 39) |
| GPQA Diamond | 89.3% (#28 of 37) | 94% (#9 of 37) (better) |
| FrontierMath Tiers 1-3 (v2) | 44.9% (#31 of 37) | 66% (#23 of 37) (better) |
| SimpleQA Verified | 30.2% (#30 of 34) | 49.3% (#19 of 34) (better) |
| Provider | xAI | xAI |
| Release date | March 10, 2026 | August 12, 2026 |
| Access | API only | API only |
| Context window | 1,000,000 tokens | 500,000 tokens |
| Max output | Not reported | Not reported |
| Input | text, image | text, image |
| Output | text | text |
When to choose each
Grok 4.20
Best for
- Comparing single-agent and multi-agent reasoning on long-context workloads
Watch out for
- The published base rates double when requests reach the provider's 200,000-token long-context threshold
Grok 4.6
Best for
- Coding, agentic, and knowledge-work API evaluations with controlled long-context cost
Watch out for
- Published token rates increase for all tokens in a request once its prompt reaches xAI's 200,000-token long-context threshold
Grok 4.20 vs Grok 4.6 FAQ
Which is cheaper, Grok 4.20 or Grok 4.6?
Grok 4.20 is 38% cheaper per 1M input tokens than Grok 4.6, $1.25 vs $2. For 10M input and 2M output tokens, Grok 4.20 costs $17.50 and Grok 4.6 $32 at standard rates.
Which scores higher on independent benchmarks, Grok 4.20 or Grok 4.6?
Grok 4.20 leads on Arena Text, 1472 Elo to 1454 Elo; Grok 4.6 leads on GPQA Diamond, 94% to 89.3%; Grok 4.6 leads on FrontierMath Tiers 1-3 (v2), 66% to 44.9%. Scores are independent evaluations, not either vendor's launch figures.
Which has the larger context window, Grok 4.20 or Grok 4.6?
Grok 4.20 has the larger context window, 1,000,000 vs 500,000 tokens.
Related comparisons
Production evaluation
Choosing between Grok 4.20 and Grok 4.6?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty