/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

OpenAI announces 25% to 50% lower GPT-3.5 Turbo prices, a GPT-4 Turbo preview model to reduce cases of “laziness”, improved text embedding models, and more

From these updates, embeddings seem like a big thing. … Forums: Msmash / Slashdot : OpenAI Drops Prices and Fixes ‘Lazy’ GPT-4 That Refused To Work

TechCrunch Devin Coldewey

Context & Ripple Effects

OpenAI had already been widening its API ladder: function calling and an earlier GPT-3.5 price cut made the lower-cost model more practical for application workflows, while fine-tuning gave developers a paid route to tailor it to their data.

The later GPT-4 Turbo launch pushed context length, structured output and token economics forward. This update extends that pattern across generation and retrieval: lower operating cost is paired with attempts to make model behavior more dependable.

First-order effects

  • GPT-3.5 Turbo API customers receive an immediate 25% to 50% price reduction, lowering the cost of existing text-generation workloads and changing the economics of workloads still being evaluated for deployment.
  • Developers gain a GPT-4 Turbo preview aimed at reducing unreliable non-completion behavior, alongside improved embedding models for retrieval and semantic-search pipelines.

Second-order effects

  • Lower GPT-3.5 Turbo pricing raises the price-performance bar for rival API providers and gives OpenAI customers more reason to route routine tasks to its cheaper tier while reserving higher-end models for harder work.
  • Better embeddings and more reliable generation can improve the end-to-end usefulness of retrieval-based applications; buyers will assess model providers on completed task quality, not token price alone.

Third-order effects

  • If repeated, these releases make model APIs less differentiated by a single flagship model and more by a continuously refreshed stack of price, reliability, customization and retrieval capabilities.
  • Falling unit prices can increase buyer leverage, but the eventual savings depend on whether reliability improvements reduce retries, fallback models and engineering overhead.

The trend: This is one data point in the shift from headline model capability toward competition on the cost per useful, reliably completed AI task.

Discussion

  • @openaidevs @openaidevs on x
    Introducing two more model updates: • Updated GPT-4 Turbo preview, completing tasks like code generation more thoroughly and fixes a bug with UTF-8 encoding • New GPT-3.5 Turbo next week, with input prices reduced by 50% and output prices reduced by 25% https://openai.com/...
  • @film_girl Christina Warren on x
    This is awesome!!
  • @gregkamradt Greg Kamradt on x
    I absolutely love how model price reductions are material GPT-3.5 Pricing (per 1K tokens): * Mar '23: $0.002 * Jul '23: $0.0015 (-25%) * Nov '23: $0.001 (-33%) * Jan '24: $0.0005 (-50%) GPT 3.5 4x cheaper than 10 months ago
  • @officiallogank @officiallogank on x
    Great news for @OpenAIDevs, we are launching: - Embedding V3 models (small & large) - Updated GPT-4 Turbo preview - Updated GPT-3.5 Turbo (*next week + with 50% price cut on Input tokens / 25% price cut on output tokens) - Scoped API keys https://openai.com/...
  • @sherwinwu Sherwin Wu on x
    It's always a great day when we can launch models that are both better and cheaper for developers - 5x cheaper embeddings, a better GPT-4 Turbo, and cheaper gpt-3.5-turbo next week!🎁🎁🎁
  • @gdb Greg Brockman on x
    Launching improved GPT models in the API, as well better and cheaper embedding models (e.g. one model is 5x cheaper than our previous embedding models with stronger performance):
  • @openaidevs @openaidevs on x
    Announcing a new generation of embedding models: • text-embedding-3-small: 5x cheaper and stronger performance compared to the previous generation • text-embedding-3-large: our best performing model, creating embeddings with up to 3072 dimensions https://openai.com/... [image]
  • @altryne Alex Volkov on x
    Open AI just addressed the “lazyness” of GPT4-Turbo with a new model gpt-4-0125-preview + released a bunch of embedding updates 👇 New embeddings model that's a bit better but 5x cheaper just dropped from open AI, say bye bye to text-embedding-ada-002 and say hello to... [image]