OpenAI vs Gemini API pricing
Updated 2026-10-18. Prices from official pricing pages, quality ranks from LMArena.
Google is cheaper at the top end right now: Gemini 4 Argon lists at $2 input and $10 output per 1M tokens, vs $10 / $50 for GPT-6 Astra (from Dec 1, Argon goes up to $4 / $20). OpenAI is cheaper at the small end: GPT-6 Luna is $0.10 / $0.50 vs Gemini 3.5 Flash-Lite at $0.30 / $2.50. On LMArena, Google's best model ranks higher for chat, creative writing and reasoning; OpenAI's ranks higher for coding, agents, vision and long documents.
| Model | Input | Output | 30M in + 6M out |
|---|---|---|---|
| GPT-6 Astra | $10 | $50 | $600 |
| GPT-5.5 | $5 | $30 | $330 |
| GPT-5.6 Sol | $4 | $20 | $240 |
| GPT-5.6 Terra | $2 | $12 | $132 |
| GPT-6.1 Sol | $2 | $10 | $120 |
| GPT-6 Sol | $2 | $10 | $120 |
| GPT-6 Luna | $0.10 | $0.50 | $6 |
| Gemini 3.1 Pro Preview | $2 | $12 | $132 |
| Gemini 4 Argon | $2 | $10 | $120 |
| Gemini 3.8 Flash | $0.75 | $3.75 | $45 |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 | $24 |
USD per 1M tokens, paid tier, short-context rate, checked 2026-10-18. Ranks are positions among the models we track on LMArena (CC BY 4.0) Text Arena and Code Arena.
Tier by tier
- Flagship: GPT-6 Astra $10 / $50 vs Gemini 4 Argon $2 / $10. Sample month (30M input, 6M output): $600 vs $120, so Astra costs 5.0x more. At today's price Argon also costs the same as GPT-6.1 Sol.
- Workhorse: GPT-6.1 Sol $2 / $10 vs Gemini 3.8 Flash $0.75 / $3.75. Sample month: $120 vs $45. Sol is #4 for coding among tracked models, Flash #14.
- Small and cheap: GPT-6 Luna $0.10 / $0.50 vs Gemini 3.5 Flash-Lite $0.30 / $2.50. Sample month: $6 vs $24.
Who ranks higher, task by task
Best model from each provider on the LMArena leaderboards we track (lower is better):
- Chat: Gemini 4 Argon #1 vs GPT-5.6 Sol #9
- Creative writing: Gemini 4 Argon #2 vs GPT-5.6 Sol #9
- Reasoning: Gemini 4 Argon #1 vs GPT-5.5 #8
- Coding: GPT-6 Astra #2 vs Gemini 4 Argon #8
- Agents: GPT-6 Astra #2 vs Gemini 4 Argon #5
- Vision: GPT-6.1 Sol #2 vs Gemini 3.8 Flash #5
- Long documents: GPT-5.6 Sol #3 vs Gemini 3.1 Pro Preview #8
Launch prices that will change
- Gemini 4 Argon: $2 in, $10 out now; $4 in, $20 out from Dec 1, per the official pricing page.
- Gemini 3.8 Flash: $0.75 in, $3.75 out now; $1.50 in, $7.50 out from Jan 1, 2027, per the official pricing page.
Things the headline price hides
- Free tier: Google's Gemini API has a free tier with usage limits for some models (Gemini 3.8 Flash is free of charge there); Gemini 4 Argon is paid only.
- Long prompts: OpenAI charges a higher rate above 272K prompt tokens on GPT-6 models, and Gemini 3.1 Pro Preview charges more above 200K.
- Thinking tokens: Google's output price includes thinking tokens, so reasoning-heavy calls cost more than the reply length suggests.
Other providers are often cheaper still: see the cheapest LLM API ranking, OpenAI vs Anthropic API pricing, or the full AI API price table. Paying for a chat app instead? Google AI Pro vs ChatGPT Plus compares the $20 plans.
Sources
FAQ
Is the Gemini API cheaper than OpenAI?
At the top end, yes for now: Gemini 4 Argon is $2 input and $10 output per 1M tokens vs $10 and $50 for GPT-6 Astra. At the small end OpenAI is cheaper: GPT-6 Luna costs $0.10 / $0.50 vs $0.30 / $2.50 for Gemini 3.5 Flash-Lite.
How much does the Gemini 4 Argon API cost?
$2 input and $10 output per 1M tokens through the launch period; from Dec 1 it is $4 input and $20 output. It is not on the free tier.
Which is better for coding, GPT or Gemini?
On LMArena's Code Arena, GPT-6 Astra is #2 and GPT-6.1 Sol #4 among the models we track, vs Gemini 4 Argon at #8. Ranks shift, so test on your own tasks.
Which is better for chat and writing?
Gemini 4 Argon is #1 on Text Arena and #2 for creative writing among tracked models; OpenAI's best are #9 and #9.
What is the cheapest OpenAI and Google model?
GPT-6 Luna ($0.10 / $0.50) and Gemini 3.5 Flash-Lite ($0.30 / $2.50) per 1M tokens, as of 2026-10-18.
Does the Gemini API have a free tier?
Yes, for some models, with usage limits, per Google's pricing page. Gemini 3.8 Flash is free of charge on the free tier; Gemini 4 Argon is paid only.
Related
- OpenAI API pricing
- Google Gemini API pricing
- OpenAI vs Anthropic API pricing
- Cheapest LLM API
- Gemini 4 Argon usage limits
- Best model per task
More guides
- Is Gemini Ultra worth it?
- Gemini limits explained
- Gemini usage limit reached: what it means and what to do
- DeepSeek vs Claude API cost
- How to check your remaining AI usage
- Is ChatGPT Plus worth it?
Not sure a plan beats the API for you? Try the plan vs API cost calculator.