GPT-4o (2024-08-06)
OpenAI · released Aug 6, 2024
- Type
- Proprietary
- Input
- $2.50
- Output
- $10.00
- Cached
- $1.25
List price · per 1M tokens · OpenAI at 128K context · source ↗
Our take
Written Aug 3, 2026GPT-4o is a multimodal model from OpenAI that accepts text, images and files, released in August 2024 with a 128,000-token request limit. Its pricing is identical across every provider we track, though benchmark scores show a slight downward drift in recent evaluations.
Choose this when you need consistent multimodal input — text, images and files — with text output, and want identical pricing no matter which provider you pick. It suits Azure-hosted deployments where 53 tokens per second is fast enough. Skip it if you need the latest coding performance, want to self-host, or need throughput guarantees without provider shopping.
The case for it
- Identical pricing across all three tracked providers — zero variance to arbitrage.
- Broad multimodal input: text, images and files in, text out.
- 128,000-token request limit handles document-heavy workflows.
The case against it
- Benchmark scores drifted slightly downward across all three Arena categories we track between July and August 2026.
- Throughput varies sharply between providers with no price compensation — one host delivers over two and a half times the speed of another at the same rate.
- LiveCodeBench score leaves most coding problems unsolved.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)110th of 143 · 1335.3
CodingWriting and fixing code on its own
Arena Coding115th of 143 · 1360.4
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored GPT-4o (2024-08-06) for this. We would take the rating from Arena Agent (IPS).
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 GPT-4o (2024-08-06) 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 model7 scoresEvery figure we hold, from 7 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
Why this differs from the header. The strip above quotes OpenAI's own list price. This is the cheapest live offer at the widest standard context we hold, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.
- per 1M tokens
- $2.50 in / $10.00 out
- Context served
- 128K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $2.50 / $10.00 | 128K | not measured | Unknown | Unknown | Unknown |
| Microsoft Azure AI | $2.50 / $10.00 | 128K | 28 tok/s | No | No | Confirmed |
| OpenAI | $2.50 / $10.00 | 128K16K out | 34 tok/s | No | Yesunknown period | Unknown |
Microsoft Azure AIAcross the 3 listings we hold: 2 say they do not train on prompts, 0 say they do and 1 do not say. 1 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 |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AI | ✓ | ✓ | ✓ |
| OpenAI | ✓ | ✓ | ✓ |
Tool calling: 3 of 3 listings say yes. JSON output: 3 of 3 listings say yes. Strict schema: 3 of 3 listings say yes.
Models people weigh against GPT-4o (2024-08-06)
When we formed this view
Dates behind this page
Prices last checked 9d ago
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 3 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->text
- Catalogue slug
- openai-gpt-4o-2024-08-06