Skip to content
Tier B — Production
Runs in:USMade in:United States
xAI

Grok 4.20 Reasoning (0309)

Tier B — Production · 1.048576M tokens

Tokonomix Editorial Team·Reviewed by Mes Kalkan··
Section 01

Speed analysis

Latency measured across all benchmark runs. P50 (median) and P95 (95th percentile) give a realistic picture of response speed under normal and peak load.

P50 latency (median)P95 latency119 runs
771261044496287812610-2311-22ms
Section 02

Pricing

What you pay per million tokens when you use this model on Tokonomix, plus an estimate for a typical conversation.

💰
API rates — Grok 4.20 Reasoning (0309)
$1.87 per 1M input tokens
$3.74 per 1M output tokens
≈ $0.0019 per typical conversation (800 tokens)
Input vs output price (per 1M tokens)
per 1M input tokens$1.87
per 1M output tokens$3.74
Section 03

Tokens per second

Throughput in tokens per second, derived from measured P50 latency. Higher is better; fluctuations track provider-side load.

Throughput (tokens / s)233 / avg 159
25752

Estimated from P50 latency × 200 output tokens — the absolute number depends on this assumption; the trend is what matters.

Section 04

Capabilities

jsontoolsvisionreasoning
Section 05

Availability

Availability

No measurements yet

We haven't recorded enough API calls to show availability stats for this model. Data appears once the model starts receiving live traffic.

Section 06

Tokonomix benchmark verdicts

● 2026-11-22

Grok 4.20 Reasoning maintains capabilities with no benchmark data updates

Grok 4.20 Reasoning (0309) shows no measurable changes in this benchmark window, with all previously established capabilities remaining stable. The model continues to support JSON, tools, vision, and reasoning features that were added in the previous period. Without new performance data, the model's standing across coding, mathematics, instruction following, and multimodal tasks remains unchanged from prior assessments. Users can expect consistent behavior with the same capability profile that was present in the last benchmark window. The lack of benchmark updates suggests either stable model performance or an absence of new evaluation data during this period. For teams relying on this model, the current window provides no evidence of regression or improvement in core competencies. Organizations should continue monitoring for future benchmark releases to track any performance shifts. The model maintains its position as a reasoning-focused offering with multimodal support, though comparative standing against other models cannot be assessed without fresh data points.

Quality

—

Latency p50

—

Test runs

0

✓ Stable capability profile maintained✗ No benchmark data updates available
Last automated test
Nov 22, 2026 · 08:04 UTC · Speed benchmark
P50 latency
857 ms
P95 latency
865 ms
Errors
0 / 6 runs
Last reviewed by Tokonomix Team·November 22, 2026