Gemini 3.5 Pro API Release Watch: What Google Has Confirmed
gemini-3.5-pro model ID. This release-watch page tracks what has been confirmed, what remains unconfirmed, and how developers can prepare without assuming an unreleased model will behave a certain way.The important point for production teams is simple: do not build plans around a speculative model name. Watch the official model list, pricing page, release notes, and Vertex/Google model pages before adding Gemini 3.5 Pro to an application roadmap.
TL;DR
- Gemini 3.5 Pro is not listed in Google's official Gemini API model docs as of May 18, 2026.
- No official
gemini-3.5-promodel ID, pricing row, context window, region list, or API launch note is currently confirmed in the checked Google docs. - Google's current Gemini 3 API documentation lists Gemini 3.1 Pro Preview, Gemini 3 Flash, Gemini 3.1 Flash-Lite, and related Gemini 3 models instead.
- Gemini 3.1 Pro is the nearest official Pro-family baseline to monitor, but Gemini 3.5 Pro should not be described as a confirmed successor until Google says so.
- Teams preparing for future Gemini models should design for model switching, fallback, cost tracking, and workload-specific evaluation.
Current Official Status
| Item | Current status | Source to monitor |
|---|---|---|
| Official Gemini 3.5 Pro release | Not confirmed in checked Google docs | Gemini API release notes |
| Gemini API model ID | Not confirmed | Gemini API model list |
| Vertex/Google model availability | Not confirmed | Google Cloud model docs |
| Pricing | Not confirmed | Gemini API pricing |
| Context window | Not confirmed | Official model card or API docs |
| Tool calling and agent features | Not confirmed for Gemini 3.5 Pro | Official model docs and capability tables |
This does not mean Google will never release Gemini 3.5 Pro. It only means developers should not treat Gemini 3.5 Pro as an available or specified API model until Google publishes it through official channels.
What Google Currently Lists Instead
gemini-3.1-pro-preview and gemini-3.1-pro-preview-customtools. There is no corresponding official pricing row for Gemini 3.5 Pro in the checked documentation.For release-watch content, this matters because the safest baseline is the current official model family, not an assumed future naming sequence.
What Developers Should Watch Before Using Gemini 3.5 Pro
Before planning production work around any future Gemini 3.5 Pro release, verify these items from official Google sources.
1. Model ID
gemini-3.5-pro. Google could use a different suffix, preview name, dated model string, or route-specific identifier. Production applications should keep model IDs in configuration rather than hard-coding them into business logic.2. API Channel
Check whether the model appears in Gemini API, Vertex AI, Google AI Studio, or only a subset of those channels. A model can be available in one product surface before another, so "released" should always specify the channel.
3. Pricing
Wait for an official pricing row before estimating production cost. For agent workloads, also calculate cost per successful task, because retries, long context, tool calls, and failed sessions can make real spend higher than the listed token price.
4. Context Window and Output Limits
Do not infer context length from the model name. Check the official model card or API documentation for input context, output limits, cache pricing, and any thresholds that change pricing above a token boundary.
5. Tool Calling and Structured Output
For agentic applications, verify function calling, structured output, code execution, grounding, and file support from the official capability table. Then test schema adherence and error recovery with real tool calls.
6. Rate Limits and Regions
For production systems, availability is as important as raw model quality. Check rate limits, regional availability, quota rules, preview restrictions, and whether higher-throughput inference options are available.
Gemini 3.5 Pro vs Gemini 3.1 Pro: A Safe Comparison Framework
Gemini 3.1 Pro is currently an official Pro-family model listed by Google, so it is the practical baseline to monitor. Gemini 3.5 Pro should be compared against it only after Google publishes official details.
| Dimension | Current Gemini 3.1 Pro signal | What to verify if Gemini 3.5 Pro launches |
|---|---|---|
| Model ID | Official model IDs are documented | Exact API model ID and channel support |
| Pricing | Official pricing row exists | Input, output, cache, batch, flex, and priority pricing |
| Context | Official model docs describe supported context | Whether context window, output limit, and pricing tiers changed |
| Agent workflows | Google positions 3.1 Pro around agentic and coding workloads | Whether task completion rate improves on real agent workloads |
| Production readiness | Preview status and rate limits apply | GA or preview status, quotas, regions, and deprecation terms |
Avoid publishing statements like "Gemini 3.5 Pro will be better for coding agents" until Google provides model details or you have your own benchmark data after release.
How to Prepare Without Depending on an Unreleased Model
Teams can prepare for future Gemini releases without making unsupported claims.
Keep Model Selection Configurable
Store model IDs, routing rules, and provider-specific options outside application logic. This makes it easier to test a new model when it appears without refactoring product code.
Track Cost Per Successful Task
For coding agents, RAG pipelines, and long-context assistants, token price is only part of the cost. Track input tokens, output tokens, retries, latency, error rate, and final task outcome.
Build Fallback Paths
Do not depend on one model for production traffic. A fallback path lets you route around outages, quota pressure, latency spikes, or model-specific regressions.
Test With Real Workloads
When a new model becomes available, benchmark it on your actual prompts, tools, documents, and acceptance criteria. Public benchmarks are useful signals, but they rarely predict every production workflow.
Separate Release Tracking From Recommendations
Before release, the article should answer "what has Google confirmed?" After release, it can become an evaluation guide with pricing, context, latency, quality, and migration recommendations.
Using EvoLink for Future Gemini Model Evaluation
EvoLink provides a unified API layer for comparing and managing multiple model families. For teams watching future Gemini models, this can reduce integration overhead and make it easier to evaluate model switching, routing behavior, and workload-level cost across providers.
Once Gemini 3.5 Pro appears in supported upstream channels, this page can be updated with the exact model ID, pricing notes, availability status, and production evaluation results.
Related articles
- Gemini 3.5 Flash API Release Watch - continue the release-watch cluster
- Gemini 3.5 Pro vs Flash Release Watch - continue the release-watch cluster
Official Sources to Monitor
FAQ
Is Gemini 3.5 Pro available in the API?
gemini-3.5-pro.What is the current official Pro-family baseline?
Gemini 3.1 Pro is the relevant official Pro-family baseline in the checked Google docs. It has documented API model IDs and pricing, while Gemini 3.5 Pro does not.
Will Gemini 3.5 Pro have a Gemini API or Vertex AI route?
That is not confirmed. It is reasonable to monitor Gemini API and Vertex/Google model documentation because previous Gemini models are documented there, but the channel, model ID, and availability terms should not be treated as confirmed until Google publishes them.
What should developers do now?
Use current official models for production planning, keep model selection configurable, track cost and reliability at the task level, and monitor Google release notes before adding Gemini 3.5 Pro to an integration roadmap.
Should this article be updated after release?
Yes. If Google releases Gemini 3.5 Pro, update this page with the exact launch date, model ID, API channel, pricing, context window, rate limits, official source links, and measured production guidance.
