AI Resources
OpenAI vs Gemini API pricing
OpenAI and Gemini API prices side by side: model tiers, current models, the cost of the same workload and what each API supports, read from the providers' own pricing pages.
Last verified: October 13, 2026
OpenAIGPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna cover reasoning, coding and high-volume work, with separate image and voice models.
GeminiSelected general-purpose models offer a free tier, multimodal input and discounted Batch and Flex token rates. Gemini 2.5 access is limited to prior active users.
Tier by tier
The closest model of each provider at four levels, shown as input / output price per million tokens.
| Tier | OpenAI | Gemini |
|---|---|---|
| Flagship | GPT-6 Astra$10/$50 | 3.1 Pro Preview$2/$12 |
| Mid | GPT-6.1 Sol$2/$10 | 3.8 Flash$0.75/$3.75 |
| Budget | GPT-6 Luna$0.10/$0.50 | 3.5 Flash-Lite$0.30/$2.50 |
| Reasoning | GPT-6 Astra$10/$50 | 3.1 Pro Preview$2/$12 |
Cost of the same workload
Base token-cost estimate for 10 equal requests: 100K input + 10K output tokens each (1M + 100K total). Only models with documented limits covering each request are shown. DeepSeek uses peak rates; its cached row assumes all input hits an existing cache, excluding warm-up misses. Gemini 2.5 requires prior active use. Cache writes, storage, tools, media and service premiums are excluded.
| OpenAI | Cost |
|---|---|
| GPT-6 Luna | $0.15 |
| GPT-5.6 Luna | $0.32 |
| GPT-5.4 Nano | $0.33 |
| GPT-5.4 Mini | $1.20 |
| GPT-6.1 Sol | $3.00 |
| GPT-6 Sol | $3.00 |
| GPT-5.6 Terra | $3.20 |
| GPT-5.6 Sol | $6.00 |
| GPT-6 Astra | $15.00 |
| Gemini | Cost |
|---|---|
| Gemini 2.5 Flash-Lite | $0.14 |
| Gemini 3.1 Flash-Lite | $0.40 |
| Gemini 3.5 Flash-Lite | $0.55 |
| Gemini 3 Flash Preview | $0.80 |
| Gemini 3.6 Flash | $1.13 |
| Gemini 3.8 Flash | $1.13 |
| Gemini 3.1 Pro Preview | $3.20 |
OpenAI models and prices
| Model | API ID | Input | Cached | Output | Context | Max out |
|---|---|---|---|---|---|---|
| GPT-6 Astra | gpt-6-astra | $10.00 | $1.00 | $50.00 | 1.05M | 128K |
| GPT-6.1 Sol | gpt-6.1-sol | $2.00 | $0.10 | $10.00 | 1.05M | 128K |
| GPT-6 Sol | gpt-6-sol | $2.00 | $0.20 | $10.00 | 1.05M | 128K |
| GPT-6 Luna | gpt-6-luna | $0.10 | $0.01 | $0.50 | 1.05M | 128K |
| GPT-5.6 Sol | gpt-5.6-sol | $4.00 | $0.40 | $20.00 | 1.05M | 128K |
| GPT-5.6 Terra | gpt-5.6-terra | $2.00 | $0.20 | $12.00 | 1.05M | 128K |
| GPT-5.6 Luna | gpt-5.6-luna | $0.20 | $0.02 | $1.20 | 1.05M | 128K |
| GPT-5.5 | gpt-5.5 | $5.00 | $0.50 | $30.00 | 1.05M | 128K |
| GPT-5.5 Pro | gpt-5.5-pro | $30.00 | N/A | $180.00 | 1.05M | 128K |
| GPT-5.4 | gpt-5.4 | $2.50 | $0.25 | $15.00 | 1.05M | 128K |
| GPT-5.4 Pro | gpt-5.4-pro | $30.00 | N/A | $180.00 | 1.05M | 128K |
| GPT-5.4 Mini | gpt-5.4-mini | $0.75 | $0.075 | $4.50 | 400K | 128K |
GPT-6 prices shown are Standard rates through 272K input tokens. Above that threshold, the full request uses 2x input/cache and 1.5x output rates. GPT-5.6 and later cache writes cost 1.25x uncached input instead of an additive fee; GPT-6.1 Sol reads cost 5% of input and the other listed GPT-6 models use 10%, with implicit or explicit caching from 1,024 tokens. Earlier-model minima vary; Pro rows do not list a cache discount. GPT-6 Batch/Flex cost 50% of Standard and Fast mode 2x. GPT-5.6 Sol pricing is promotional through at least November 21, 2026. Check individual older-model long-context rules. The 2x long-context multiplier also applies to cache writes. The fine-tuning platform is closed to new users. Astra Ultrafast short-context rates are $60 input, $6 cached input, $75 cache writes and $300 output per million tokens. GPT-5.4 Nano and GPT-5.3 Codex are deprecated, with shutdown scheduled for 2027-04-01.
Gemini models and prices
| Model | API ID | Input | Output | Cached | Context | Free |
|---|---|---|---|---|---|---|
| 3.8 Flash | gemini-3.8-flash | $0.75 | $3.75 | $0.075 | 1M | yes |
| 3.6 Flash | gemini-3.6-flash | $0.75 | $3.75 | $0.075 | 1M | yes |
| 3.5 Flash-Lite | gemini-3.5-flash-lite | $0.30 | $2.50 | $0.03 | 1M | yes |
| 3.1 Flash-Lite | gemini-3.1-flash-lite | $0.25 | $1.50 | $0.025 | 1M | yes |
| 3.1 Pro Preview | gemini-3.1-pro-preview | $2.00 / $4.00 | $12.00 / $18.00 | $0.20 / $0.40 | 1M | no |
| 3 Flash Preview | gemini-3-flash-preview | $0.50 | $3.00 | $0.05 | 1M | yes |
Selected models, not the complete Google catalog. Text-input and text-output rates are shown; audio and media can differ. Gemini 3.6/3.8 Flash standard input/output is $0.75/$3.75 per million tokens through 2026-12-31, then $1.50/$7.50 from 2027-01-01. The listed general-purpose text models have a 1,048,576-token input limit and 65,536-token output limit. Gemini 2.5 Pro, 2.5 Flash and 2.5 Flash-Lite require prior active use and are not deprecated. This restriction also applies to the $0.10/$0.40 Flash-Lite option in charts, simulation and the input KPI. Free tiers depend on model and eligibility; 3.1 Pro Preview has none. Explicit caching adds storage charges; implicit caching is enabled by default on Gemini 2.5 and newer. Audio input/cache rates per million tokens are $0.50/$0.05 on 3.1 Flash-Lite, $1.00/$0.10 on 3 Flash Preview and 2.5 Flash, and $0.30/$0.03 on 2.5 Flash-Lite. 3.5 Flash-Lite uses $0.30 input and $0.03 cache for all its input modalities. Imagen 4 is retired; its row is historical. The earliest possible shutdown date for the three Veo 3.1 preview variants is 2026-10-22; Google recommends gemini-omni-1.1-flash. They are not yet shut down.
What each API supports
| Capability | OpenAI | Gemini |
|---|---|---|
| Vision (image in) | Yes | Yes |
| Video input | Not documented | Native |
| Audio input | Realtime 2.1 | Native |
| PDF input | PDF text and page images | Native bytes |
| Image generation | GPT Image 2.5 | Nano Banana 2 |
| Video generation | Videos API shut down 2026-09-24 | Gemini Omni 1.1 Flash |
| Text to speech | Realtime 2.1 Mini | 3.8 Flash TTS |
| Embeddings | text-embedding-3-small | Embedding 2 |
| Web search | $10/1K | 3.x: 5K/month free, then $14/1K search queries |
| Tool / function calling | Yes | Yes |
| Structured JSON | Schema enforced | Schema enforced |
| Reasoning mode | Effort levels | Thinking tokens |
| Code fill-in-the-middle | Not documented | Not documented |
| Prompt caching | GPT-6: 90–95% off reads; 1.25x writes | Implicit; explicit storage billed hourly |
| Batch API | Yes, 50% off | Yes, 50% off |
| Long-context surcharge | GPT-6 above 272K: 2x input/cache/write, 1.5x output | 2x input / 1.5x output above 200K (Pro) |
| Free tier | Not documented | Selected text models; 2.5 access restricted |
| Max context | 1.05M | 1M |
| Max output | 128K | 64K |
| OpenAI SDK compatible | Native | OpenAI libraries via compatibility endpoint |
| Data sovereignty | Regional (1.1x) | Not documented |
Scheduled price changes
Dated prices recorded in the pricing data, per million tokens (input / output).
Sources
Every figure on this page was read from the provider's own pricing documentation on the verification date shown above.