Model Tracker

GPT-4 Turbo

GPT-4 Turbo is an AI model made by OpenAI. OpenAI released it on 6 November 2023. It is deprecated. It takes text, image and returns text.

OpenAI

GPT-4 Turbo is an AI model made by OpenAI. OpenAI released it on 6 November 2023. It is deprecated. It takes text, image and returns text. Its context window is 128,000 tokens. Its longest answer is 4,096 tokens.

Released: 6 November 2023 — openai.com, read 19 October 2026
Status: Deprecated — developers.openai.com, read 19 October 2026
Takes: Text, Image — developers.openai.com, read 19 October 2026
Returns: Text — developers.openai.com, read 19 October 2026
Open weights: Not confirmed: The maker's pages reviewed do not confirm whether GPT-4 Turbo has open weights.
Licence: Not confirmed: The maker's pages reviewed do not state a licence for GPT-4 Turbo.
Context window: 128,000 tokens — developers.openai.com, read 19 October 2026
Longest answer: 4,096 tokens — developers.openai.com, read 19 October 2026
Knowledge cutoff: Not confirmed: The maker's pages disagree. The model page says “Dec 01, 2023 knowledge cutoff” (GPT-4 Turbo Model page); the announcement says “GPT‑4 Turbo is more capable and has knowledge of world events up to April 2023.” (DevDay announcement).
API price, input: Not confirmed: The maker's current API pricing page does not list GPT-4 Turbo, so its standard input price cannot be confirmed from the required pricing page.
API price, output: Not confirmed: The maker's current API pricing page does not list GPT-4 Turbo, so its standard output price cannot be confirmed from the required pricing page.
Where: API — developers.openai.com, read 19 October 2026
Parameters: Not confirmed: The maker's pages reviewed do not publish a parameter count for GPT-4 Turbo.
Model card: Not confirmed: The maker's pages reviewed do not provide a GPT-4 Turbo model card or system card.

Benchmarks

Measured by others, not by the maker: each score with the name the measurer gives the model and the setting it ran, its day and its licence.

Arena

Text Arena, gpt-4-turbo-2024-04-09: 1,325 (1,321–1,328, 98,114 votes), published 15 October 2026
Vision Arena, gpt-4-turbo-2024-04-09: 1,112 (1,100–1,123, 13,391 votes), published 9 October 2026

Arena’s leaderboard dataset (lmarena-ai/leaderboard-dataset on Hugging Face), CC BY 4.0: ratings from people’s votes between two anonymous models, as Arena publishes them. Data · CC BY 4.0

Epoch AI

Epoch Capabilities Index, GPT-4 Turbo (Apr 2024): 127.3 (121.6–128.8), as of 19 October 2026
Epoch Capabilities Index, GPT-4 Turbo (Nov 2023): 126.5 (121.1–129.2), as of 19 October 2026
GPQA Diamond, gpt-4-turbo-2024-04-09: 46.6% (± 2.7), run 27 February 2025
GPQA Diamond, gpt-4-1106-preview: 42.4% (± 2.4), run 27 January 2025
GPQA Diamond, gpt-4-0125-preview: 42.3% (± 2.7), run 27 January 2025
FrontierMath, tiers 1–3, gpt-4-turbo-2024-04-09: 0.7% (± 0.5), run 27 August 2026
OTIS Mock AIME 2024–2025, gpt-4-turbo-2024-04-09: 6.7% (± 2.4), run 27 February 2025

Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks, CC BY 4.0: Epoch AI’s own runs and its Capabilities Index. Data · CC BY 4.0

Questions

When was GPT-4 Turbo released?

OpenAI released GPT-4 Turbo on 6 November 2023.

Is GPT-4 Turbo still available?

It is deprecated: OpenAI has announced the date it will retire it.

What is the context window of GPT-4 Turbo?

The context window of GPT-4 Turbo is 128,000 tokens, as OpenAI prints it.

What does GPT-4 Turbo take and return?

GPT-4 Turbo takes text, image and returns text.

Sources

GPT-4 Turbo Model | OpenAI API (read 19 October 2026)
Deprecations | OpenAI API (read 19 October 2026)
New models and developer products announced at DevDay (read 19 October 2026)
Pricing | OpenAI API (read 19 October 2026)

All of the Model Tracker