ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

Grok 4.20 ReasoningvsGrok 4.3

Grok 4.20 Reasoning leads 3 of 5 shared benchmarks. Both list the same combined token price. Both offer a 1M-token context window.

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
3 – 2
Grok 4.20 Reasoning leads
Cheaper per token
Tie
list price, input + output
Larger context
Tie
both 1M tokens
Provider uptime (30d)
xAI
Grok 4.20 Reasoning
xAI · released 2026-03-09
textvision
Context
1M
Max out
Input /1M
$1.25
Output /1M
$2.5
Cached /1M
$0.2
Scores
5 · 5 core
textvision
Context
1M
Max out
Input /1M
$1.25
Output /1M
$2.5
Cached /1M
$0.2
Scores
6 · 6 core
Quality

Benchmark matrix

BenchmarkGrok 4.20 ReasoningGrok 4.3ΔEdge
Reported by both · 5
GPQA Diamond89.3% epoch88.8% epoch+0.5 ptGrok 4.20 Reasoning
FrontierMath Tier 4 v2 (Epoch AI run)17.1% epoch14.6% epoch+2.5 ptGrok 4.20 Reasoning
FrontierMath Tiers 1-3 v2 (Epoch AI run)44.9% epoch42.8% epoch+2.1 ptGrok 4.20 Reasoning
OTIS Mock AIME 2024-2025 (Epoch AI run)92.2% epoch93.3% epoch-1.1 ptGrok 4.3
SimpleQA Verified30.2% epoch33.2% epoch-3 ptGrok 4.3
Only Grok 4.20 Reasoning reports · 0
Grok 4.20 Reasoning reports nothing Grok 4.3 does not.
Only Grok 4.3 reports · 1
LMArena Elonot reported1397.7
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Grok 4.20 Reasoning minus Grok 4.3 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecGrok 4.20 ReasoningGrok 4.3Edge
Context window1M tokens1M tokensTie
Max output
Input price / 1M$1.25 $1.25 Tie
Output price / 1M$2.5 $2.5 Tie
Cached input / 1M$0.2 $0.2 Tie
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$3.75$3.75Tie
Modalitiestext · visiontext · visionTie
Released2026-03-09
Cited benchmark scores56
Reliability

xAI status

All providers
More matchups

Grok 4.20 Reasoning vs …

More matchups

Grok 4.3 vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run Grok 4.20 Reasoning and Grok 4.3 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.