RETIRED131K CONTEXTARCHIVE PRICECOST-EFFICIENT REASONING
Grok 3 Mini API Pricing
Grok 3 Mini is an archive row for xAI's small 2025 reasoning model: $0.30/M input and $0.50/M output. The launch page positioned it as cost-efficient reasoning; current traffic should move to Grok 4.3.
Input - per 1M tokens
$0.30/M
Historical Grok 3 Mini archive
Output - per 1M tokens
$0.50/M
No current endpoint archive
Cache N/A - billed as input
$0.30/M
No cache row listed N/A
Effective - agentic blend
$0.32/M
92/8 split - 82% cache
§ 01 / TERMINAL
Run the numbers.
Calculator pre-loaded with Grok 3 Mini archive rates. The model is retired, so use this for invoice replay and migration math.
$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS
Real-world presets.
ARCHIVE
Invoice replay
$0.300/1M in
AGENT
Legacy agent task
$0.028/task
MIGRATION
Migration cost comparison
$0.065/pack
RAG
Archived RAG answer
$0.051/answer
§ 03 / TAPE
Price history.
Input · $0.30/M
Output · $0.50/M
Cached · $0.30/M
FEB 17 Launched at $0.3/M input and $0.5/M outputMAY 18 Archive status verified against current xAI docs and launch page
§ 04 / TOKENIZER
Paste text. See tokens. See cost.
Estimate · grok-tokenizer-estimate · ≈3.85 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 —
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Grok 3 Mini Current | $0.30 | $0.50 | $0.32 archive 92/8 | 131K | Archive cheap Grok reasoning |
| Grok 4.3 | $1.25 cache $0.20 | $2.50 | $0.56 current replacement | 1M | Current Grok default |
| Gemini 2.5 Flash | $0.30 cache $0.03 | $2.50 | $0.27 budget Gemini | 1M | Low-cost multimodal work |
| GPT-5.4 mini | $0.75 cache $0.07 | $4.50 | $0.54 OpenAI mini | 400K | Subagents and lightweight coding |
| DeepSeek V4 Pro | $0.43 cache $0.00 | $0.87 | $0.14 budget reasoning | 1M | Low-cost reasoning alternative |
Frequently asked.
Grok 3 Mini pricing questions, with archive status separated from token math.
Q · 01 What was Grok 3 Mini's API price? +
This archive page uses
$0.3/M input, and $0.5/M output. Use it for historical billing and migration math, not fresh production routing.Q · 02 Is Grok 3 Mini still available? +
No. This is an archive row for a retired model family; xAI now directs text workloads to
grok-4.3.Q · 03 What should replace it? +
Use
grok-4.3. xAI's current pricing page lists Grok 4.3 at $1.25/M input, $0.20/M cached input, and $2.50/M output.Q · 04 Does prompt caching apply? +
No separate cached-input price is listed in this archive row. The calculator treats cached input as regular input at
$0.3/M.Q · 05 Why keep an archive page? +
Retired model prices still matter for old invoices, migration plans, benchmarks, and generation-to-generation price history. The page is marked as archive so it does not imply current endpoint availability.
Q · 06 How accurate is the tokenizer estimate? +
The widget uses
4.875 characters per token as a planning estimate. Exact billing can vary with language, hidden reasoning, cached prompt boundaries, and tool usage.