Headless Thrones

OpenAI · Official-source model record

GPT-3.5 Turbo · gpt-3.5-turbo-instruct

GPT · Version gpt-3.5-turbo-instruct

RetiredChecked Oct 23, 2026
Editorial (0) ↓Download model brief (.md)

Publisher-sourced inventory. Specifications come from official documentation and model cards. “Not documented” means we have no verified value. This inventory is separate from our editorial reviews.

Coverage & sources ↗

Current catalog answers

What to know about GPT-3.5 Turbo · gpt-3.5-turbo-instruct

AI-compiled from linked model documentation. This is a catalog reference, not an independent performance review. Catalog checked October 23, 2026.

What happened to GPT-3.5 Turbo · gpt-3.5-turbo-instruct, and what replaces it?

These details come from the linked publisher material for this model. Reported capabilities and training descriptions are publisher statements, not independent test results.

API shutdown notice
OpenAI lists 2026-09-28 as the shutdown date for this exact snapshot. This is lifecycle documentation, not an independent live availability test. Source
Recommended migration
The current deprecation table recommends gpt-5.6-terra; evaluate that model separately rather than attributing its specifications to this snapshot. Source
Historical listed pricing
The exact model page lists $1.50 per million input tokens and $2 per million output tokens. The retirement notice takes precedence for availability; these are historical/reference rates. Source

Which GPT-3.5 Turbo · gpt-3.5-turbo-instruct does this page describe?

GPT-3.5 Turbo · gpt-3.5-turbo-instruct is a model record from OpenAI in the GPT family. Source

The recorded model identifier or checkpoint is gpt-3.5-turbo-instruct. Use that exact identifier when checking documentation or a provider listing. Source

How can you access GPT-3.5 Turbo · gpt-3.5-turbo-instruct?

The catalog records the historical API access route for GPT-3.5 Turbo · gpt-3.5-turbo-instruct. Source

As of the catalog check, this version was marked retired. Source Check the linked lifecycle information before starting a new integration.

For a new integration, use the publisher’s replacement guidance and check its current account, region and endpoint requirements. Historical API access does not mean the retired endpoint can still be called.

What are GPT-3.5 Turbo · gpt-3.5-turbo-instruct's context and output limits?

The documented context window is 4,096 tokens. Source

The documented maximum output is 4,096 tokens. Source

Context, input and output limits describe different constraints. Do not add the figures together to infer a larger window. Check the endpoint's accounting for conversation history, system instructions, tools, media and generated tokens; a documented maximum is not a guarantee of answer quality or a measured usable capacity.

What dates are documented for GPT-3.5 Turbo · gpt-3.5-turbo-instruct?

Knowledge cutoff: September 1, 2021. Source This labels the documented knowledge boundary; it is not a promise that every earlier fact is known or correct.

Retrieval or web tools can supply newer information at run time. That does not change the model's recorded training or knowledge dates, and this record does not establish that a particular endpoint includes those tools.

What can GPT-3.5 Turbo · gpt-3.5-turbo-instruct receive and generate?

Documented inputs: text. Source

Documented outputs: text. Source

Input support and output generation are separate capabilities. Reading an image does not establish image generation; a text output listing does not establish speech output. File sizes, formats and media limits depend on the documented implementation.

Keep the exact version in view

Other GPT models from OpenAI.

These are separate records in the same publisher family. A shared family name does not establish equal capabilities or interchangeable licenses.

Chat Latest

API · 400,000 context tokens

GPT-5.2

API · 400,000 context tokens

GPT-5.2 Pro

API · 400,000 context tokens

GPT-5.4

API · 1,050,000 context tokens

GPT-5.4 Mini

API · 400,000 context tokens

GPT-5.4 Pro

API · 1,050,000 context tokens

The documented record

Specifications, with their sources.

A missing value remains unknown. Input limits, output limits and the total context window are separate measurements.

Publisher
OpenAISource ↗
Family
GPTSource ↗
Version
gpt-3.5-turbo-instructSource ↗
Availability
RetiredSource ↗
Access
APISource ↗
Inputs
TextSource ↗
Outputs
TextSource ↗
Context window
4,096 tokensSource ↗
Maximum input
Not documented
Maximum output
4,096 tokensSource ↗
Knowledge cutoff
Sep 1, 2021Knowledge cutoff · day precisionSource ↗
Training-data cutoff
Not documented
Training completed
Not documented
Release date
Not documented
License
Not documented
Input price / 1M tokens
Not documented
Output price / 1M tokens
Not documented

Read this with the numbers.

Exact snapshot listed on the vendor family page. Family-default specifications are not assumed for this older snapshot. Vendor API shutdown date: 2026-09-28.

Model identifiers and aliases 0

No additional identifier has been verified.

Follow the evidence

Official sources.

  1. deprecations — official OpenAI documentation ↗developers.openai.com
  2. gpt 3.5 turbo — official OpenAI documentation ↗developers.openai.com
  3. gpt 3.5 turbo instruct.md — official OpenAI documentation ↗developers.openai.com

From the Headless Thrones newsroom

Editorial about GPT-3.5 Turbo · gpt-3.5-turbo-instruct

No published articles are linked to this exact model yet. New coverage will appear here as it is published.

Your starting point

GPT-3.5 Turbo · gpt-3.5-turbo-instruct

OpenAI · API

Browse all models ↗
Inputs
TextSource ↗
Outputs
TextSource ↗
Context window
4,096 tokensSource ↗
Knowledge cutoff
Sep 1, 2021Knowledge cutoff · day precisionSource ↗

A place to begin comparing

55 models to consider.

Matches must share an access route, support every documented input type of your starting model, and return text. When both publish a context limit in the same field, the candidate must fall between half and twice that limit. Unknown or differently defined limits remain as broader matches, after verified near-context matches. Within each group, recent documented releases appear first. These criteria do not establish equal quality. The full current inventory is considered.

Cohere

command-a-translate-08-2025

Text input; text output · Shared access: API · Documented context within ½–2×

Cohere

tiny-aya-earth

Text input; text output · Shared access: API · Documented context within ½–2×

Cohere

tiny-aya-fire

Text input; text output · Shared access: API · Documented context within ½–2×

Cohere

tiny-aya-global

Text input; text output · Shared access: API · Documented context within ½–2×

Cohere

tiny-aya-water

Text input; text output · Shared access: API · Documented context within ½–2×

Tencent

hy-mt2-lite

Text input; text output · Shared access: API · Documented context within ½–2×

Tencent

hy-mt2-plus

Text input; text output · Shared access: API · Documented context within ½–2×

Tencent

hy-mt2-pro

Text input; text output · Shared access: API · Documented context within ½–2×

Microsoft

MAI-1-preview

Text input; text output · Shared access: API · Capacity match not established

Google

Gemini 2.5 Flash

Text input; text output · Shared access: API · Capacity match not established

Google

Gemini 2.5 Flash Lite

Text input; text output · Shared access: API · Capacity match not established

Google

Gemini 2.5 Pro

Text input; text output · Shared access: API · Capacity match not established

A record you can follow

The source belongs beside the fact.

Official sources

Model records cite the developer’s documentation, release notes and model cards. A listing alone is not a tested service or a recommendation.

Dates mean different things

Knowledge cutoff, training-data cutoff, training completion and release date are separate fields. A model’s name or repository upload date does not establish any of them.

Limits need context

Some publishers use rounded K/M token figures. Deployment, reasoning settings and regional access can change usable limits. Read the source before choosing.

Coverage and last check

Independent editorial model library assembled from primary vendor documentation and official publisher model cards. This is a dated, auditable inventory of the sources reviewed, not a claim to every model worldwide. First-party language-model inventory for OpenAI, Anthropic, Google Gemini/Gemma, xAI, Meta and Reflection. Includes vendor-listed legacy/retired identities with explicit statuses; official pretrained and instruction-tuned checkpoints remain distinct. Excludes embeddings, moderation/guard models, image/audio/video-only generation, community fine-tunes, speculative draft helpers, format/runtime/quantization duplicates, and PaliGemma dataset-specific benchmark fine-tunes. Availability means documented vendor API or downloadable official weights, not guaranteed account eligibility. Unknown specifications and dates remain null. Rounded K/M limits preserve vendor granularity; exact integers are used only where documented. No OpenRouter facts are used. Primary-source inventory of language-capable models from Alibaba/Qwen, DeepSeek, Moonshot, MiniMax, Z.ai, ByteDance Seed, Tencent, Baidu, Xiaomi, StepFun, 01.AI, Shanghai AI Laboratory, Baichuan and OpenBMB. Includes full reachable official public weight inventories plus documented hosted API models and explicit retirements. Embeddings, reward/classifier/router/tokenizer/draft-head components, standalone OCR/document converters and image/video/audio-only generators are excluded; multimodal language models retained. Official packaging variants are folded into aliases where a canonical listing was identifiable; remaining research/base/SFT variants retain exact identities. Not a guarantee every historical/private/regional model is available. Unknown specifications remain null; K/M API display quantities use decimal normalization where explicitly noted, rather than implying verified exact serving ceilings. Model-name dates and repository creation timestamps are never used as release/training dates. Open weights does not mean unrestricted or OSI-open-source licensing. Some model-card bodies were unreachable; their public inventory metadata is retained and disclosed. Official Mistral, Cohere, AI21, NVIDIA Nemotron, Microsoft Phi/MAI, Amazon Nova, IBM Granite, AllenAI OLMo/Molmo, TII Falcon, Nous Research, Arcee and Essential AI model inventory checked 2026-10-23. Includes base, instruction, reasoning, coding and text-producing multimodal models, including legacy API versions. Excludes embeddings/rerankers/reward and safety classifiers, image/audio/video-only generation, robotics-only models, duplicate quantization/runtime packages, adapters, intermediate checkpoint experiments, and third-party mirrors. Arcee coverage focuses on its named product families rather than exploratory merge repositories. Repository listing means listed, not independently verified API availability. K/M figures are nominal decimal normalizations when the source supplies shorthand; deployment-specific exact token limits may differ. Unverified values remain null. This is a source-backed inventory, not a complete claim-level independent editorial review. Supplemental official Liquid AI, Reka and Inception generative language catalog. Excludes media-only generators and decision/classification-only models. No assumed training dates.

Catalog edition updated . Source links are provided on each model record.

View the full coverage report and discovery sources ↗