High-volume, cost-sensitive work
Choose Grok 4.20 Non-Reasoning
Grok 4.20 Non-Reasoning costs 19% less on a typical mix of three input tokens for every output token ($1.56 against $1.93 per million tokens).
Grok 4.20 Non-Reasoning vs o4-mini: price, context and capabilities
Choose Grok 4.20 Non-Reasoning for high-volume, cost-sensitive work, writing and code generation, reused prompts and documents (RAG, agents), very long inputs, tool use and support agents and the latest release. Choose o4-mini for reasoning over long documents.
In short
- Grok 4.20 Non-Reasoning is 19% cheaper than o4-mini at the headline price ($1.56 / 1M tokens vs $1.93 / 1M tokens).
- Grok 4.20 Non-Reasoning reads more per request: 1M tokens against 200k.
- Grok 4.20 Non-Reasoning is the newer model (March 2026, against Apr 16, 2025).
- Only o4-mini offers reasoning.
Which should you choose?
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Writing and code generation
Choose Grok 4.20 Non-Reasoning
Grok 4.20 Non-Reasoning charges $2.50 per million output tokens, against $4.40 for o4-mini. Output dominates the bill when answers are long.
Reused prompts and documents (RAG, agents)
Choose Grok 4.20 Non-Reasoning
Grok 4.20 Non-Reasoning charges $0.20 per million cached input tokens, against $0.28 for o4-mini.
Very long inputs
Choose Grok 4.20 Non-Reasoning
Grok 4.20 Non-Reasoning reads up to 1M tokens per request, against 200k for o4-mini.
Agentic coding
Either one
They are practically level on Terminal-Bench Hard (16.7% and 15.2%).
Tool use and support agents
Choose Grok 4.20 Non-Reasoning
Grok 4.20 Non-Reasoning scores 59.9% on τ²-Bench (agentic tool use), against 55.6% for o4-mini.
Reasoning over long documents
Choose o4-mini
o4-mini scores 61% on Long-context reasoning (AA-LCR), against 23.3% for Grok 4.20 Non-Reasoning.
Latest release
Choose Grok 4.20 Non-Reasoning
Grok 4.20 Non-Reasoning is the newer release (March 2026, against Apr 16, 2025 for o4-mini).
Monthly cost at three volumes
| Volume | Grok 4.20 Non-Reasoning | o4-mini | Difference |
|---|---|---|---|
| Prototype: 1M input, 250K output tokens | $1.88 | $2.20 | Grok 4.20 Non-Reasoning, $0.33 less |
| Production: 20M input, 5M output tokens | $37.50 | $44.00 | Grok 4.20 Non-Reasoning, $6.50 less |
| Scale: 500M input, 100M output tokens | $875.00 | $990.00 | Grok 4.20 Non-Reasoning, $115.00 less |
Standard rates, without caching or batch discounts. Volumes are per month.
Specifications side by side
| Grok 4.20 Non-Reasoning | o4-mini | |
|---|---|---|
| Provider | xAI | OpenAI |
| Category | Language models | Language models |
| Status | Generally available | Deprecated |
| Headline price | $1.56 / 1M tokens | $1.93 / 1M tokens |
| Input | $1.25 / 1M tokens | $1.10 / 1M tokens |
| Output | $2.50 / 1M tokens | $4.40 / 1M tokens |
| Cached input | $0.20 / 1M tokens | $0.275 / 1M tokens |
| Context window | 1M | 200k |
| Max output | No data | 100k |
| Released | March 2026 | Apr 16, 2025 |
| Knowledge cutoff | No data | June 2024 |
| Licence | Proprietary | Proprietary |
| Capabilities | Vision, Tool calling, Structured output, Prompt caching, Batch API | Vision, Reasoning, Tool calling, Structured output, Prompt caching, Batch API |
| API model id | grok-4.20-0309-non-reasoning | o4-mini |
Benchmarks side by side
Leader in bold.
| Benchmark | Grok 4.20 Non-Reasoning | o4-mini |
|---|---|---|
| Terminal-Bench Hard | 16.7% | 15.2% |
| GPQA Diamond | 77.6% | 78.4% |
| Humanity's Last Exam | 27.9% | 16.5% |
| IFBench (instruction following) | 49.3% | 68.7% |
| τ²-Bench (agentic tool use) | 59.9% | 55.6% |
| MMMU-Pro (visual reasoning) | 64.9% | 69.2% |
| Long-context reasoning (AA-LCR) | 23.3% | 61% |
Independent evaluations by Artificial Analysis, run the same way for every model, at the reasoning setting shown. Percentages unless marked as an index.
Frequently asked questions
Should I choose Grok 4.20 Non-Reasoning or o4-mini?
Choose Grok 4.20 Non-Reasoning for high-volume, cost-sensitive work, writing and code generation, reused prompts and documents (RAG, agents), very long inputs, tool use and support agents and the latest release. Choose o4-mini for reasoning over long documents.
Which is cheaper, Grok 4.20 Non-Reasoning or o4-mini?
Grok 4.20 Non-Reasoning is 19% cheaper than o4-mini at the headline price ($1.56 / 1M tokens vs $1.93 / 1M tokens).
Which has the larger context window, Grok 4.20 Non-Reasoning or o4-mini?
Grok 4.20 Non-Reasoning reads more per request: 1M tokens against 200k.
What is the difference between Grok 4.20 Non-Reasoning and o4-mini?
Grok 4.20 Non-Reasoning is made by xAI and o4-mini by OpenAI. Grok 4.20 Non-Reasoning is 19% cheaper than o4-mini at the headline price ($1.56 / 1M tokens vs $1.93 / 1M tokens). Grok 4.20 Non-Reasoning reads more per request: 1M tokens against 200k. Grok 4.20 Non-Reasoning is the newer model (March 2026, against Apr 16, 2025). Only o4-mini offers reasoning.
Which is better for coding, Grok 4.20 Non-Reasoning or o4-mini?
They are practically level on Terminal-Bench Hard (16.7% and 15.2%).
How much do Grok 4.20 Non-Reasoning and o4-mini cost per month?
At 20 million input and 5 million output tokens a month, Grok 4.20 Non-Reasoning costs $37.50 and o4-mini costs $44.00, at standard rates.
Which is newer, Grok 4.20 Non-Reasoning or o4-mini?
Grok 4.20 Non-Reasoning was released on March 2026 and o4-mini on Apr 16, 2025.