/
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

xAI launches an API for Grok 3 and Grok 3 Mini with “reasoning” capabilities; Grok 3 costs $3/$15 per million input/output tokens, Grok 3 Mini costs $0.30/$0.50

Kyle Wiggers / TechCrunch :

TechCrunch Kyle Wiggers

Context & Ripple Effects

xAI’s path moved from a Grok-2 beta whose API availability was planned to a Grok-3 beta for X Premium+ users. This release turns the newer Grok 3 and smaller Grok 3 Mini models into products developers can invoke programmatically.

The split between a higher-priced Grok 3 and far cheaper Grok 3 Mini makes the launch as much a packaging and usage-pricing decision as a model release. It creates a clear cost/performance choice for application builders.

First-order effects

  • Developers can now integrate Grok 3 and Grok 3 Mini through an API rather than relying on access through X subscriptions; Grok 3 is priced at $3 per million input tokens and $15 per million output tokens.
  • The $0.30/$0.50-per-million-token pricing for Grok 3 Mini gives customers a substantially lower-cost option for workloads that do not require the flagship model.

Second-order effects

  • Teams adopting Grok must route tasks between the two models based on output quality and token costs, making workload design central to the effective price of reasoning features.
  • The published rates give buyers a direct benchmark for evaluating competing reasoning-model APIs and increase pressure on providers to differentiate on both capability and inference cost.

Third-order effects

  • If providers continue pairing reasoning models with sharply differentiated API tiers, model selection will increasingly be an application-level unit-economics decision rather than a single-vendor subscription choice.
  • This points toward reasoning capability becoming commercially standardized through metered APIs, with durable advantage depending on the cost of useful outputs rather than model access alone.

The trend: Reasoning models are shifting from subscriber-facing launches into tiered, metered API products where developers optimize capability against token economics.