Last verified
VERIFIED PRICELOW-LATENCY SNAPSHOT1MTEXT MODELPROMPT CACHING

Grok 4.20 (0309) Non-Reasoning API Pricing

Grok 4.20 0309 Non-Reasoning is a dated latency-focused xAI snapshot at $1.25/$2.50. $1.25/M input, $2.5/M output, and $0.2/M cached input. Use it to compare non-reasoning Grok routing against reasoning and multi-agent variants.

Input - per 1M tokens
$1.25/M
Source xAI active
Output - per 1M tokens
$2.50/M
Use for cost checks active
Cached input
$0.20/M
Cache hit price discounted
Effective - agentic blend
$0.56/M
92/8 split - 82% cache
§ 01 / TERMINAL

Run the numbers.

Calculator pre-loaded with Grok 4.20 (0309) Non-Reasoning rates. Tweak spend, output mix, or cache hit rate to compare this model with current alternatives.

$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS

Real-world presets.

AGENT LOOP

Repo-wide bug fix

$0.049/task
81k in - 7k out~2,040 units/$100
LONG DOC

Reading 100-page contracts

$0.088/doc
175k in - 8k out~1,136 units/$100
RAG SUPPORT

Support agent ticket triage

$0.004/ticket
4k in - 1k out~25,000 units/$100
ASSISTANT TURN

Research planning turn

$0.010/turn
12k in - 2k out~10,000 units/$100
§ 03 / TAPE

Price history.

Grok 4.20 (0309) Non-Reasoning is active; archived at $1.25/$2.5 per M.

Input · $1.3/M
Output · $2.5/M
Cached · $0.20/M
MAR 09 Dated non-reasoning snapshot at $1.25/M input, $0.20/M cached input, and $2.50/M outputMAY 18 Active dated Grok 4.20 variant verified on xAI model docs
§ 04 / TOKENIZER

Paste text. See tokens. See cost.

Estimate · grok-tokenizer-estimate · ≈4.2 chars/token Auto-counts as you type

This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.

Characters —
Words —
Tokens (estimated) —
Cost as input · uncached —
Cost as output · uncached —
Cost as cached input —
§ 05 / SHELF

Up against the shelf.

All models →
Model Input /M Output /M Effective blended Context Best for
Grok 4.20 (0309) Non-Reasoning $1.25 cache $0.20 $2.50 $0.56 current page 1M Grok long-context agents
Grok 4.3 $1.25 cache $0.20 $2.50 $0.56 same blend 1M Grok long-context agents
Grok 4 $3.00 cache $0.75 $15.00 $2.26 pricier 256K Grok long-context agents
Grok 4 Fast $0.20 cache $0.05 $0.50 $0.11 cheaper 2M Grok long-context agents
GPT-5.4 $2.50 cache $0.25 $15.00 $1.80 pricier 1.05M Current OpenAI replacement
Claude Sonnet 4.6 $3.00 cache $0.30 $15.00 $1.92 pricier 1M Current Claude agents
Gemini 2.5 Pro $1.25 cache $0.13 $10.00 $1.10 pricier 2M Gemini multimodal workloads

Frequently asked.

Short answers for teams checking Grok 4.20 (0309) Non-Reasoning pricing, status, and migration choices.

Q · 01 Is Grok 4.20 (0309) Non-Reasoning still available? +
Grok 4.20 (0309) Non-Reasoning is listed as active in the current snapshot and was checked against the vendor model/pricing page.
Q · 02 How much does Grok 4.20 (0309) Non-Reasoning cost? +
Grok 4.20 (0309) Non-Reasoning is listed at $1.25/M input and $2.5/M output.
Q · 03 Is cached-input pricing included? +
Yes. Cached input is shown at $0.2/M.
Q · 04 What should teams compare it against? +
Use the shelf to compare current replacements and nearby models. Snapshot replacement: current same-provider models.
Q · 05 Why keep this page? +
These pages preserve price history, preview economics, and migration context without pretending every older or preview model is the best current default.
Y
Reviewed by Yaroslav Vikhariev Founder - AI//COST - Last verified 2026-05-18 - Vendor pricing
Methodology Report a correction More by Y.V.