Compare / Grok 4.20 vs Grok 4.3

Grok 4.20 vs Grok 4.3

Which is cheaper, which is more capable, and which does better at each kind of work, from 8 results on 2 sources that measured both models.

xAIGrok 4.20 Released 17 Feb 2026 · $1.25 in · $2.50 out
xAIGrok 4.3 Released 17 Apr 2026 · $1.25 in · $2.50 out
Grok 4.20 wins 31 too close to callGrok 4.3 wins 4

Of the 8 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee).

Grok 4.20 or Grok 4.3? The short answer

They score and cost about the same: pick on the results below that match your work.

  • They cost about the same: $1.56 and $1.56 per million tokens (three input to one output).
  • They score about the same on the Artificial Analysis Intelligence Index (25.7 and 24.9).
  • Grok 4.3 is newer: released 17 Apr 2026, against 17 Feb 2026 for Grok 4.20.
  • On third-party benchmarks, on the results both models have: Grok 4.20 does better for agents and tool use and reasoning and knowledge.

Head to head

25.7
Intelligence Index
24.9
$1.56
Price per million tokens, blendedlower is better
$1.56
2,000,000
Context window, tokens
1,000,000

Every fact side by side

Grok 4.20Grok 4.3
Price per million tokens$1.25 in · $2.50 out$1.25 in · $2.50 out
Blended (3 in : 1 out)$1.56$1.56
Cheapest host, blended$1.56 (xAI)$1.56 (xAI)
Cached input, per million–$0.20 cached input
Intelligence Index25.724.9
Spring Prompt overall––
Output speed, tokens a second––
Context window2,000,000 tokens1,000,000 tokens
Released17 Feb 202617 Apr 2026 (newer)
WeightsClosedClosed
Developer based inthe USthe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 84 of 159 117 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects not measured 49 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 61 of 187 62 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality not measured 58 of 169
Sticking to the factsSummarising without inventing things, and factual answers not measured 107 of 139

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureGrok 4.20Grok 4.3
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 76 · 82 of 249 measured 25.7 24.9high reasoning
GPQA Diamond% of questions, higher is better 31 · 41 of 232 measured 91.1% 90.1%high reasoning
Humanity's Last Exam% of questions, higher is better 69 · 59 of 250 measured 34.5% 37.2%high reasoning
Vending-Bench 2 reported by Vending-Bench 2 · 1 result
Money after a yearUS dollars, higher is better 29 · 53 of 58 measured $4,663 $35.26
Show all 8 results (4 more)
MeasureGrok 4.20Grok 4.3
Artificial Analysis reported by Artificial Analysis · 4 results
IFBench% of instructions, higher is better 4 · 1 of 195 measured 81.2% 83.3%medium reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 123 · 85 of 241 measured 69.0% 75.0%medium reasoning
Terminal-Bench Hard% of tasks, higher is better 39 · 39 of 190 measured 37.9% 37.9%high reasoning
τ²-bench telecom% of tasks, higher is better 32 · 7 of 193 measured 93.0% 97.7%high reasoning

Quick answers

Is Grok 4.20 better than Grok 4.3?

They score and cost about the same: pick on the results below that match your work.

Which is cheaper, Grok 4.20 or Grok 4.3?

They cost about the same: $1.56 and $1.56 per million tokens (three input to one output).

Which is better for agents and tool use, Grok 4.20 or Grok 4.3?

Grok 4.20 ranks 84 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 117 for Grok 4.3.

Which is better for reasoning and knowledge, Grok 4.20 or Grok 4.3?

Grok 4.20 ranks 61 of 187 models on sale for hard questions across science, maths and general knowledge, against 62 for Grok 4.3.

Which has the bigger context window, Grok 4.20 or Grok 4.3?

Grok 4.20: 2,000,000 tokens, against 1,000,000 for Grok 4.3.

Run Grok 4.20 and Grok 4.3 on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.