Compare / head-to-head
Grok 4.20 ReasoningvsGrok 4 Fast Reasoning
Grok 4.20 Reasoning leads 1 of 1 shared benchmark. Grok 4 Fast Reasoning is 5.4x cheaper per token. Grok 4 Fast Reasoning has the larger context window (2M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 0
Grok 4.20 Reasoning leads
Cheaper per token
Grok 4 Fast Reasoning
5.4x cheaper, input + output
Larger context
Grok 4 Fast Reasoning
2M tokens
Provider uptime (30d)
—
xAI
Grok 4.20 Reasoning
xAI · released 2026-03-09
textvision
Grok 4 Fast Reasoning
xAI · released 2025-09-19
text
Quality
Benchmark matrix
1 shared · 4 only Grok 4.20 Reasoning · 11 only Grok 4 Fast Reasoning| Benchmark | Grok 4.20 Reasoning | Grok 4 Fast Reasoning | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 1 | ||||
| GPQA Diamond | 89.3% ↗epoch | 85.7% ↗ | +3.6 pt | Grok 4.20 Reasoning |
| Only Grok 4.20 Reasoning reports · 4 | ||||
| FrontierMath Tier 4 v2 (Epoch AI run) | 17.1% ↗epoch | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 44.9% ↗epoch | not reported | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 92.2% ↗epoch | not reported | — | — |
| SimpleQA Verified | 30.2% ↗epoch | not reported | — | — |
| Only Grok 4 Fast Reasoning reports · 11 | ||||
| AIME 2025 | not reported | 92% ↗ | — | — |
| Humanity's Last Exam (no tools) | not reported | 20% ↗ | — | — |
| BrowseComp | not reported | 44.9% ↗ | — | — |
| BrowseComp (zh) | not reported | 51.2% ↗ | — | — |
| HMMT 2025 (no tools) | not reported | 93.3% ↗ | — | — |
| LiveCodeBench (Jan-May) | not reported | 80% ↗ | — | — |
| Reka Research Eval | not reported | 66% ↗ | — | — |
| Search Arena Elo | not reported | 1163 ↗ | — | — |
| SimpleQA | not reported | 95% ↗ | — | — |
| X Bench Deepsearch (zh) | not reported | 74% ↗ | — | — |
| X Browse | not reported | 58% ↗ | — | — |
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 Fast Reasoning 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
Official pricing pages and model cards| Spec | Grok 4.20 Reasoning | Grok 4 Fast Reasoning | Edge |
|---|---|---|---|
| Context window | 1M tokens | 2M tokens | Grok 4 Fast Reasoning |
| Max output | — | — | — |
| Input price / 1M | $1.25 ↗ | $0.2 ↗ | Grok 4 Fast Reasoning |
| Output price / 1M | $2.5 ↗ | $0.5 ↗ | Grok 4 Fast Reasoning |
| Cached input / 1M | $0.2 ↗ | $0.05 ↗ | Grok 4 Fast Reasoning |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $3.75 | $0.7 | Grok 4 Fast Reasoning |
| Modalities | text · vision | text | Grok 4.20 Reasoning |
| Released | 2026-03-09 | 2025-09-19 | — |
| Cited benchmark scores | 5 | 12 | — |
Reliability
xAI status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
Grok 4.20 Reasoning vs …
Models sharing the most benchmarksMore matchups
Grok 4 Fast Reasoning vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Grok 4.20 Reasoning and Grok 4 Fast Reasoning on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.