text model comparison
Grok 4.20 vs Grok 4.3
Grok 4.20 and Grok 4.3 are both current text models, from xAI. Both cost $1.25 per 1M input tokens. On Arena Text, Grok 4.20 scores 1472 Elo and Grok 4.3 1442 Elo (Arena, measured October 8, 2026). Both accept 1,000,000 tokens of context.
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 4.3 |
|---|---|---|
| Price per 1M input tokens | $1.25 | $1.25 |
| Price per 1M output tokens | $2.50 | $2.50 |
| Cost of 10M input and 2M output tokens | $17.50 | $17.50 |
| Arena Text | 1472 Elo (#22 of 39) (better) | 1442 Elo (#33 of 39) |
| GPQA Diamond | 89.3% (#28 of 37) (better) | 88.8% (#29 of 37) |
| FrontierMath Tiers 1-3 (v2) | 44.9% (#31 of 37) (better) | 42.8% (#32 of 37) |
| SimpleQA Verified | 30.2% (#30 of 34) | 33.2% (#28 of 34) (better) |
| Provider | xAI | xAI |
| Release date | March 10, 2026 | Not reported |
| Access | API only | API only |
| Context window | 1,000,000 tokens | 1,000,000 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 4.3
Best for
- General-purpose reasoning, instruction following, and agentic tool-use evaluations
Watch out for
- The published base rates double when requests reach the provider's 200,000-token long-context threshold
Grok 4.20 vs Grok 4.3 FAQ
Which is cheaper, Grok 4.20 or Grok 4.3?
Both cost $1.25 per 1M input tokens. For 10M input and 2M output tokens, Grok 4.20 costs $17.50 and Grok 4.3 $17.50 at standard rates.
Which scores higher on independent benchmarks, Grok 4.20 or Grok 4.3?
Grok 4.20 leads on Arena Text, 1472 Elo to 1442 Elo; Grok 4.20 leads on GPQA Diamond, 89.3% to 88.8%; Grok 4.20 leads on FrontierMath Tiers 1-3 (v2), 44.9% to 42.8%. Scores are independent evaluations, not either vendor's launch figures.
Which has the larger context window, Grok 4.20 or Grok 4.3?
Both accept 1,000,000 tokens of context.
Related comparisons
Production evaluation
Choosing between Grok 4.20 and Grok 4.3?
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