Gemini 2.5 Pro
Google · released Jun 17, 2025
- Type
- Proprietary
- Input
- $1.25
- Output
- $10.00
- Cached
- Not published
List price · per 1M tokens · Google AI at 1M context · source ↗ · a reseller below undercuts it; the table carries the spread
Our take
Written Aug 2, 2026Gemini 2.5 Pro is Google's flagship reasoning model that accepts text, images, files, audio and video in a single request, with a one-million-token limit. It offers tiered throughput pricing and scores well across coding, reasoning and creative writing benchmarks, though it is available only through Google's hosted services.
Choose this for long-context multimodal work where you need to feed documents, video or audio into the same model, or for latency-sensitive coding and reasoning where 94.5–99 tokens per second matters. The slowest tier suits budget-sensitive batch work. Skip it if you need to self-host, fine-tune, or avoid proprietary lock-in; also skip if web-development coding is your main need, where it lags its own general coding score by a wide margin.
The case for it
- Broad multimodal input coverage: text, images, files, audio and video in a single model.
- One-million-token request limit, disclosed as 1,048,576 tokens.
- Strong measured performance across diverse tasks: Arena Coding 1465.2, Arena Hard Prompts 1458.3, Arena Creative Writing 1444.3 — all within a 21-point spread suggesting balanced capability.
- Tiered pricing allows throughput-cost tradeoff: the same mid-tier rate is available at 99, 96 or 94.5 tokens per second, or half the price at 26–85 tokens per second.
The case against it
- Proprietary lock-in with no local or fine-tuning path: weights are proprietary, licence undisclosed, parameter count unpublished.
- Premium tier pricing is steep: 3.6 times the entry input price and 3.6 times the entry output price.
- Web development coding lags other coding measures: Arena Code (WebDev) 1224.2 versus Arena Coding 1465.2, a 241-point gap.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)39th of 143 · 1445.5
CodingWriting and fixing code on its own
Arena Coding58th of 143 · 1465.2
AgenticPlanning, calling tools, staying on task
Gemini 2.5 Pro is not on Arena Agent (IPS), which is where the rating would come from, so there is no rating here. It is on SWE-bench Verified, in 24th of 39 with 53.6.
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Gemini 2.5 Pro placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Or rent it from someone else
The lab is also the cheapest we hold. The strip above and this offer are the same one, so nothing on this page undercuts Google AI on 1M of context.
- per 1M tokens
- $1.25 in / $10.00 out
- Context served
- 1M
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Google AI Studioflex tier | $0.63 / $5.00 | 1M | 20 tok/s | No | Yes55 days | Unknown |
| Google Vertex AIglobal | $0.63 / $5.00 | 1M | 85 tok/s | No | No | Confirmed |
| Google AI | $1.25 / $10.00 | 1M66K out | not measured | Unknown | Unknown | Unknown |
| OpenRouter | $1.25 / $10.00 | 1M | not measured | Unknown | Unknown | Unknown |
| Google Vertex AIus | $1.25 / $10.00 | 1M | 104 tok/s | No | No | Confirmed |
| Google AI Studio | $1.25 / $10.00 | 1M | 81 tok/s | No | Yes55 days | Unknown |
| Google Vertex AIeu | $1.25 / $10.00 | 1M | 99 tok/s | No | No | Confirmed |
| DeepInfra | $1.25 / $10.00 | 1M | not measured | Unknown | Unknown | Unknown |
| Google AI Studiopriority tier | $2.25 / $18.00 | 1M | not measured | No | Yes55 days | Unknown |
Google AI Studio
Google Vertex AI
OpenRouter
DeepInfraAcross the 9 listings we hold: 6 say they do not train on prompts, 0 say they do and 3 do not say. 3 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| Google AI Studioflex | ✓ | ✓ | ✓ |
| Google Vertex AIglobal | ✓ | ✓ | ✓ |
| Google AI | |||
| OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIus | ✓ | ✓ | ✓ |
| Google AI Studio | ✓ | ✓ | ✓ |
| Google Vertex AIeu | ✓ | ✓ | ✓ |
| DeepInfra | |||
| Google AI Studiopriority | ✓ | ✓ | ✓ |
Tool calling: 7 of 9 listings say yes, 2 publish no parameter list. JSON output: 7 of 9 listings say yes, 2 publish no parameter list. Strict schema: 7 of 9 listings say yes, 2 publish no parameter list.
Models people weigh against Gemini 2.5 Pro
When we formed this view
Dates behind this page
Prices last checked 9d ago
What we do not know about this model yet
- 2 of 9 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 3 of 9 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.
Identifiers
- Modality record
- text+image+file+audio+video->text
- Catalogue slug
- google-gemini-2-5-pro