Gemini 1.5 Pro: Complete Specifications, Pricing, API Access & Use Cases (2026)
What Is Gemini 1.5 Pro?
Gemini 1.5 Pro is Google DeepMind’s mid-size multimodal large language model, introduced for early testing on February 15, 2024, featuring a historical 2M-token maximum context window and multimodal understanding, with no current Gemini API pricing for the retired model as of June 2026. Google’s original announcement described Gemini 1.5 Pro as the first Gemini 1.5 model released for early testing, with a standard 128K-token context window and a 1M-token private preview window; Google later made a 2M-token context window generally available for Gemini 1.5 Pro on June 27, 2024.
Gemini 1.5 Pro was important because it helped make long-context multimodal workflows more practical, including long document analysis, video understanding, large codebase review, and multi-file reasoning. It is now mainly relevant as a historical and migration reference, because Google Cloud’s model lifecycle documentation lists gemini-1.5-pro-001 and gemini-1.5-pro-002 as retired model versions.
What Are Gemini 1.5 Pro’s Key Specifications and Pricing?
| Field | Gemini 1.5 Pro Value |
|---|---|
| Provider | Google DeepMind / Google (as of June 2026) |
| Model Family | Gemini 1.5 (as of June 2026) |
| Model Type | Multimodal large language model with Mixture-of-Experts architecture (as of June 2026) |
| Release Date | Early testing announced February 15, 2024; gemini-1.5-pro-001 GA noted May 23, 2024 (as of June 2026) |
| Context Window | Historical maximum 2M tokens for Gemini 1.5 Pro; original announcement listed 128K standard and 1M private preview (as of June 2026) |
| Input Pricing | No current Gemini API price confirmed because Gemini 1.5 Pro versions are retired (as of June 2026) |
| Cached Input Pricing | Historical context caching was announced for Gemini 1.5 Pro and 1.5 Flash; no current Gemini 1.5 Pro cache price applies after retirement (as of June 2026) |
| Output Pricing | No current Gemini API price confirmed because Gemini 1.5 Pro versions are retired (as of June 2026) |
| Pricing Unit | Current official pricing unit not applicable for retired Gemini 1.5 Pro endpoints (as of June 2026) |
| Modality Support | Text, code, image, audio, video, and document-style inputs historically supported; text output (as of June 2026) |
| Supported Input Types | Text, code, image, video, audio, PDFs/documents through supported Gemini API workflows during availability (as of June 2026) |
| Supported Output Types | Text output; structured output capabilities existed in Gemini API workflows, but current support for retired Gemini 1.5 Pro endpoints is not applicable (as of June 2026) |
| API Access | Historical access through Google AI Studio, Gemini API, and Vertex AI; current Gemini 1.5 Pro versions retired (as of June 2026) |
| Model ID | gemini-1.5-pro-001, gemini-1.5-pro-002; retired (as of June 2026) |
| Availability | Retired; not recommended for new production use (as of June 2026) |
| Knowledge Cutoff | Not confirmed from official sources as of June 2026. |
| Rate Limits | Historical paid-tier limit increased to 1,000 RPM in September 2024; current rate limits not applicable after retirement (as of June 2026) |
| Fine-tuning Support | Not confirmed for Gemini 1.5 Pro from official sources as of June 2026. |
| Streaming Support | Gemini API supports streaming generally, but current Gemini 1.5 Pro endpoint support is not applicable after retirement (as of June 2026). |
| Batch API Support | Not applicable for retired Gemini 1.5 Pro endpoints as of June 2026. |
| Tool / Function Calling | Gemini API added parallel function calling in May 2024, but current Gemini 1.5 Pro endpoint support is not applicable after retirement (as of June 2026). |
| Structured Output / JSON Mode | Not confirmed for current Gemini 1.5 Pro endpoints as of June 2026 because the model versions are retired. |
| License / Usage Restrictions | Governed historically by Google Gemini API / Google Cloud terms; current endpoint access is retired (as of June 2026). |
What Can Gemini 1.5 Pro Do That Makes It Useful in Production?
Gemini 1.5 Pro was useful for long-context analysis because it could process very large prompts, including long text, codebases, and multimedia inputs. Google described examples such as one hour of video, 11 hours of audio, large codebases, and hundreds of thousands of words in a single context, which made the model relevant for document review, research synthesis, and repository analysis.
It also supported multimodal understanding. Google’s examples included reasoning over video, processing long PDFs, image description, object detection, and structured extraction from visual content. These capabilities made it useful for workflows that mixed text, visual evidence, and document layout information.
For production planning, however, teams should treat Gemini 1.5 Pro as a legacy reference rather than a current default. Developers comparing long-context models or unified model routing should separate model capability from lifecycle risk, because an unavailable model cannot safely anchor a new production system.
What Are Gemini 1.5 Pro’s Supported Modalities?
| Modality | Supported? | Notes |
|---|---|---|
| Text input | Historical yes | Used for long-form prompts, document text, and general instructions. |
| Code input | Historical yes | Google cited large-codebase reasoning examples. |
| Image input | Historical yes | Used for image description, visual reasoning, object detection, and extraction. |
| Audio input | Historical yes | Google cited processing up to 11 hours of audio in long-context examples. |
| Video input | Historical yes | Google cited video understanding and long video processing. |
| PDF / document input | Historical yes | Google cited processing 1,000-page PDFs and document reasoning examples. |
| Audio output | Not confirmed | Gemini 1.5 Pro was not documented as a native speech-output model in the cited sources. |
| Image or video output | Not confirmed | It was a multimodal understanding model, not a confirmed image/video generation model. |
Where Does Gemini 1.5 Pro Fall Short?
The main limitation is lifecycle status. Google Cloud documentation lists Gemini 1.5 Pro model versions as retired, which means new systems should not depend on them for API access, pricing, rate limits, or support.
A second limitation is cost and latency at very large context sizes. Google noted during the original rollout that very long context could involve longer latency, and later announced context caching to reduce costs for repeated-token workloads.
A third limitation is general AI reliability. This is a general AI limitation and is not model-specific unless stated by the provider: outputs may contain hallucinations, missing context, or incorrect interpretations, especially in legal, medical, financial, or safety-critical tasks. Expert review remains necessary for high-stakes use.
What Is Gemini 1.5 Pro Best Used For?
| Use Case | Why Gemini 1.5 Pro May Fit | Important Limitation |
|---|---|---|
| Historical model reference | Useful for understanding the evolution of long-context multimodal AI. | Retired endpoints limit current production use. |
| Long-document analysis design | Its 2M-token context window shaped expectations for large-document workflows. | Use current supported models for deployment. |
| Multimodal workflow planning | The model demonstrated text, code, image, audio, video, and PDF understanding. | Historical capability does not guarantee current API availability. |
| Model migration audits | Helps teams identify legacy gemini-1.5-pro-* dependencies. | Current migration targets should be verified in current Google docs. |
| Competitive comparison | Useful when comparing older long-context systems with newer models such as Gemini 2.0 Flash, GPT-4o, or Claude 3.5 Sonnet. | Compare against current availability, not only historical specs. |
How Does Gemini 1.5 Pro Compare to GPT-4o and Claude 3.5 Sonnet?
| Comparison Area | Gemini 1.5 Pro | GPT-4o | Claude 3.5 Sonnet | Scenario Fit |
|---|---|---|---|---|
| Provider | Google DeepMind / Google | OpenAI | Anthropic | Choose based on verified current access and governance needs. |
| Lifecycle status | Retired Gemini 1.5 Pro versions as of June 2026. | Requires current OpenAI verification. | Requires current Anthropic verification. | Gemini 1.5 Pro is mostly a legacy comparison point. |
| Context window | Historical 2M-token maximum. | Not compared here without current source verification. | Not compared here without current source verification. | Gemini 1.5 Pro is relevant for long-context history. |
| Modality profile | Historical multimodal input understanding across text, code, image, audio, video, and documents. | Often compared for multimodal workflows; verify current details. | Often compared for writing, coding, and analysis workflows; verify current details. | Use current provider docs before choosing a production model. |
| API access | Historical Gemini API / Google AI Studio / Vertex AI; retired now. | See current OpenAI documentation or internal GPT-4o model profile. | See current Anthropic documentation or internal Claude reference. | Production decisions should prioritize active supported endpoints. |
This comparison is intentionally limited. It does not declare an overall winner because the most important difference for Gemini 1.5 Pro in 2026 is lifecycle status, not just capability.
How Do I Access Gemini 1.5 Pro?
As of June 2026, Gemini 1.5 Pro should be treated as retired for provider API access. Google Cloud’s model lifecycle documentation lists gemini-1.5-pro-001 with a May 24, 2025 retirement date and gemini-1.5-pro-002 with a September 24, 2025 retirement date; Google’s Vertex AI release notes also state that access to Gemini 1.5 models was discontinued on September 24, 2025.
This page therefore does not include executable Python or curl examples for Gemini 1.5 Pro. Code examples for retired endpoints could mislead developers into building against unavailable model IDs. Teams with legacy references to gemini-1.5-pro should verify their current Google migration path and replace retired model IDs with supported current models.
FAQs
What was Gemini 1.5 Pro’s context window?
Gemini 1.5 Pro originally launched with a 128K-token standard context window and 1M-token private preview access. Google later made a 2M-token context window generally available for Gemini 1.5 Pro in June 2024.
How much does Gemini 1.5 Pro cost?
As of June 2026, there is no current Gemini API price to cite for Gemini 1.5 Pro because its listed model versions are retired. Historical pricing changed during 2024, but current production pricing should be checked only for supported models.
Can I still use Gemini 1.5 Pro through the API?
Google Cloud documentation lists gemini-1.5-pro-001 and gemini-1.5-pro-002 as retired. New production systems should not depend on Gemini 1.5 Pro API access.
What replaced Gemini 1.5 Pro?
Google’s lifecycle documentation recommended newer Gemini models after Gemini 1.5 Pro retirement. The exact replacement depends on the API surface, region, modality, and current Google model availability.
