Model comparison
GPT-5.5 vs Gemini 3.1 Pro Preview: a flat 2.5x gap
GPT-5.5 costs 2.5x as much as Gemini 3.1 Pro Preview on every rate and every workload. Where they differ instead: output limits, long prompts, and access.
· Prices as of September 28, 2026
GPT-5.5
OpenAI · Released April 23, 2026 · Previous generation
OpenAI's April 2026 flagship for coding and professional work, now a previous generation that stays in the API.
GPT-5.5 facts and comparisonsGemini 3.1 Pro Preview
Google · Released February 19, 2026 · Preview
Google's only current Pro model, still in preview, positioned for deep reasoning and agentic coding.
Gemini 3.1 Pro Preview facts and comparisons
The short answer
Gemini 3.1 Pro Preview costs less on every line: GPT-5.5's input, output, and cache prices are all 2.5x higher, and the example agentic coding session costs $5.00 on GPT-5.5 against $2.00. GPT-5.5 is OpenAI's April 2026 flagship and leaves ChatGPT and Codex sign-in on October 14, 2026, while Gemini 3.1 Pro Preview is Google's current Pro model, still in preview. Keep GPT-5.5 for API workflows that need its 128K output limit, and choose Gemini 3.1 Pro Preview for the lower cost in Gemini CLI.
Choose GPT-5.5 if
- Your responses run past 65.5K tokens: GPT-5.5 writes up to 128K.
- You have API integrations built on GPT-5.5, which stays in the API after it leaves ChatGPT and Codex sign-in.
- OpenAI's claims fit your agents: precise tool use on large tool surfaces and long-running agent tasks.
- You rely on GitHub Copilot, which retired Gemini 3.1 Pro Preview on September 1, 2026.
Choose Gemini 3.1 Pro Preview if
- Price leads: every rate is 60% lower than GPT-5.5's.
- You use Gemini CLI, where this model makes up the Pro half of the default auto model.
- Your agent relies on custom tools, which Google says its customtools endpoint prioritizes better alongside bash.
Side by side
Specs and prices
| Fact | GPT-5.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Maker | OpenAI | |
| API model id | gpt-5.5 | gemini-3.1-pro-preview |
| Released | April 23, 2026 | February 19, 2026 |
| Status | Previous generation | Preview |
| Context window | 1.05M tokens | 1.05M tokens |
| Max output | 128K tokens | 65.5K tokens |
| Open weights | No | No |
| Input, per 1M tokens | $5 | $2 |
| Cache hit, per 1M | $0.50 | $0.20 |
| Cache write, per 1M | $5 (same as input) | $2 (same as input) |
| Output, per 1M tokens | $30 | $12 |
| Runs in | Codex, Cursor, OpenCode, OpenRouter, and GitHub Copilot | Cursor, Gemini CLI, OpenCode, and OpenRouter |
Standard API rates in US dollars, as published by each maker on September 28, 2026. Batch and priority tiers, taxes, and subscription plans are not included. GPT-5.5: Prompts over 272K input tokens cost 2x for input and 1.5x for output, for the full session. Gemini 3.1 Pro Preview: Prompts over 200K input tokens cost $4 input, $0.40 cached, and $18 output per million tokens.
Cost
What the same work costs
The same token counts, priced at each model’s published rates. Cached tokens are billed at each maker’s cache prices, so the session shows what caching is worth on each model.
Real sessions differ: the two models count the same code as different numbers of tokens, and reasoning settings change how much each one writes. Your own history is the real test.
| Workload | GPT-5.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Agentic coding session, 100K input, 400K written to cache, 2M read from cache, 50K output | $5.00 | $2.00 |
| Large one-off review, 150K input with no cache hits, 10K output | $1.05 | $0.42 |
| Output-heavy generation, 30K input, 80K output | $2.55 | $1.02 |
| A month of sessions, 110 sessions: 5 a day, 22 working days | $550.00 | $220.00 |
| Where the session’s cost goes | ||
| Cache writes | $2.00 | $0.80 |
| Cache reads | $1.00 | $0.40 |
| Uncached input | $0.50 | $0.20 |
| Output | $1.50 | $0.60 |
- caching saves on the session with GPT-5.5 (64%)
- $9.00
- caching saves on the session with Gemini 3.1 Pro Preview (64%)
- $3.60
Why every workload shows the same 2.5x
Most model comparisons have a caching twist. This one does not. GPT-5.5 charges $5 input, $0.50 per cache hit, and $30 output per million tokens, and Gemini 3.1 Pro Preview charges $2, $0.20, and $12. Every rate is 2.5x apart.
Both also bill cache writes as ordinary input: OpenAI adds no write charge on GPT-5.5 and earlier, and Google publishes no separate write price. With identical caching rules and proportional prices, every workload shows the same 2.5x: $5.00 against $2.00 for the session, $1.05 against $0.42 for the uncached review, and $2.55 against $1.02 for output-heavy generation.
The session's cost splits the same way on both: 40% cache writes, 20% cache reads, 10% input, and 30% output. Caching saves 64% on each, $9.00 on GPT-5.5 and $3.60 on Gemini. Over 110 sessions a month the totals are $550.00 and $220.00, a $330.00 difference.
Matching long-context surcharges, different triggers
Both models list a 1.05M context window, and both raise prices for long prompts by the same multiples, 2x for input and 1.5x for output. Google applies Gemini's higher tier, $4 input, $0.40 cached, and $18 output, to prompts over 200K input tokens. OpenAI applies GPT-5.5's to prompts over 272K input tokens, and for the full session.
The practical difference is the trigger. A prompt between 200K and 272K tokens pays Gemini's long-context rates but not GPT-5.5's, and a GPT-5.5 session that crosses 272K pays the higher rates for the whole session. The example session stays under 200K per request, so the tables use standard rates for both.
GPT-5.5 leaves Codex sign-in while Gemini awaits 3.5 Pro
OpenAI calls GPT-5.5 "A new class of intelligence for coding and professional work" and says it reaches strong results with fewer reasoning tokens than earlier models at the same effort. It leaves ChatGPT and Codex sign-in on October 14, 2026 and stays in the API, and Cursor, OpenCode, OpenRouter, and GitHub Copilot list it too. For complex coding the Codex docs now recommend GPT-6 Sol.
Gemini 3.1 Pro Preview has been a preview since February 19, 2026, and Google has announced Gemini 3.5 Pro without releasing it. After GitHub Copilot retired it on September 1, 2026, it remains in Gemini CLI, Cursor, OpenCode, and OpenRouter. Thinking is always on for it, at a high level by default.
Output limits are the clearest functional gap: 128K tokens per response on GPT-5.5 and 65.5K on Gemini 3.1 Pro Preview. EveryToken reads your Codex and Gemini CLI history on a Mac and prices it at API rates, which shows what each model costs across your real sessions.
Prompt caching
How each maker bills cached tokens
OpenAI
Prompt caching is on by default. From GPT-5.6 on you can also mark up to four explicit cache breakpoints, while GPT-5.5 and earlier cache automatically only.
From GPT-5.6 on, a cache write costs 1.25x the uncached input price and a cache hit costs 0.1x. GPT-5.5 and earlier add no charge for writing the cache: written tokens are billed as ordinary input, and a hit costs 0.1x on the models compared here.
On GPT-5.6 and later, a cached prefix stays reusable for at least 30 minutes after its last use, and caching starts at 1,024 input tokens.
Source: OpenAI: Prompt caching
Implicit caching is on by default for Gemini 2.5 and newer. When a request repeats a prefix Google has cached, the discount is applied automatically, but a hit is not guaranteed.
Explicit caching creates a cache you reference by name, with a guaranteed discount. It adds a storage charge for as long as the cache lives, 1 hour by default: $4.50 per million tokens per hour on Pro models and $0.50 to $1 on Flash models.
On every model compared here, a cache hit costs 10% of the input price. Google publishes no separate cache-write price, so this blog prices written tokens as ordinary input.
Source: Google: Context caching
Your own numbers
See what GPT-5.5 and Gemini 3.1 Pro Preview really cost you.
everyaitoken reads your Codex, Cursor, OpenCode, OpenRouter, and Gemini CLI history on your Mac and prices every request at API rates, with what caching saved or cost. $9 once.
FAQ
Questions
How much more expensive is GPT-5.5 than Gemini 3.1 Pro Preview?
It costs 2.5x as much on every rate and every workload here. The example agentic session costs $5.00 against $2.00, a $3.00 difference, as an API-equivalent estimate.
What happens to GPT-5.5 on October 14, 2026?
It leaves ChatGPT and Codex sign-in. Codex users who sign in with ChatGPT will need to switch models, while the model itself stays available through the API.
Do both models charge more for long prompts?
Yes. Gemini 3.1 Pro Preview moves to $4 input and $18 output above 200K input tokens. GPT-5.5 charges 2x input and 1.5x output above 272K, applied to the full session.
Is there a cache-write premium on either model?
No. GPT-5.5 bills written tokens as ordinary input, as OpenAI does for GPT-5.5 and earlier, and Google has no separate write price. GPT-5.5 caches automatically only, while Gemini also offers explicit caches with a storage charge of $4.50 per million tokens per hour on Pro models.
Sources
- OpenAI: API pricing
- OpenAI docs: GPT-5.5
- OpenAI: Using GPT-5.5
- OpenAI: API changelog
- Codex docs: Models
- Cursor docs: Models and pricing
- OpenRouter: GPT-5.5
- Google: Gemini API pricing
- Google docs: Gemini 3.1 Pro Preview
- Google: Gemini models
- Google: Gemini 3.1 Pro
- Google DeepMind: Gemini
- Gemini CLI source: model configuration
- Cursor docs: Gemini 3.1 Pro
- OpenRouter: Gemini 3.1 Pro Preview
- OpenCode docs: Zen
- GitHub Docs: Supported AI models in Copilot
- OpenAI: Prompt caching
- Google: Context caching