AI model APIs · head-to-head
Gemini 3.1 Pro vs GPT-5.6 Sol
Same usage, official rates, ranked by nothing but price. Both rate cards, the monthly bill at five workloads, and the exact usage mix where the answer flips.
The pricing verdict Gemini 3.1 Pro is cheaper than GPT-5.6 Sol at every input:output mix — there is no usage pattern where GPT-5.6 Sol costs less.
Rate cards
| Per 1M tokens | Gemini 3.1 Pro | GPT-5.6 Sol |
|---|---|---|
| Input | $2 | $5 |
| Cached input | $0.20 | $0.50 |
| Output | $12 | $30 |
| Context window | — | 1M |
Monthly cost at five workloads
| Workload | Tokens / month | Gemini 3.1 Pro | GPT-5.6 Sol |
|---|---|---|---|
| Prototype | 1M in · 0.3M out | $5.60 | $14 |
| Small app | 10M in · 3M out | $56 | $140 |
| Production app | 50M in · 15M out | $280 | $700 |
| High volume | 200M in · 60M out | $1,120 | $2,800 |
| Heavy platform | 1B in · 300M out | $5,600 | $14,000 |
Standard on-demand rates from each provider's official pricing page, verified July 2026; figures exclude caching and batch discounts and any long-context surcharges. Gemini 3.1 Pro: Rates shown are for prompts ≤200k tokens; above 200k, input doubles to $4 and output rises to $18 per 1M. Batch runs at 50%. GPT-5.6 Sol: Batch API runs at 50% of these rates. Price is one input — quality, latency and context limits differ. Confirm live pricing before committing.