Tech Meridian← ALL MODELS
PROMY MERIDIAN

XAI (SPACEXAI) · MODEL RELEASE TRACKER

Grok 4.5

Grok 4.5 is a released model from xAI (SpaceXAI) described as their most capable model for coding, agentic tasks, and knowledge work. The announcement highlights strong coding performance, multi-step agentic reasoning, fast serving speed, and broad availability across web, mobile, productivity add-ins, Grok Build, GitHub Copilot, and the SpaceXAI console.

CURRENT SNAPSHOT4/5 DIMENSIONS WITH DATA

The dimensions that change the decision.

PRICING
Input price (<200K input)$2 USD / 1M input tokens; prompt <200,000 tokens
Input price (≥200K input)$4 USD / 1M input tokens; prompt ≥200,000 tokens; all request tokens use this rate
Output price (<200K input)$6 USD / 1M output tokens; prompt <200,000 tokens
Output price (≥200K input)$12 USD / 1M output tokens; prompt ≥200,000 tokens; all request tokens use this rate
Cached input price (<200K input)$0.3 USD / 1M cached input tokens; prompt <200,000 tokens
Cached input price (≥200K input)$0.6 USD / 1M cached input tokens; prompt ≥200,000 tokens; all request tokens use this rate
Pricing (SpaceXAI console)Offered from the SpaceXAI console at $2 per million input tokens and $6 per million output tokens.
CONTEXT WINDOW
Context window500,000 tokens
MODALITIES
Input modalitiesText, images
Output modalitiesText
Modalities / task typesFocused on code and text-based knowledge work and agentic multi-step tasks (coding, science, engineering, math-related knowledge work).
BENCHMARKS

Not established from the available sources.

AVAILABILITY
Platform availabilityMade available on grok.com, X, Grok iOS and Android apps, and integrates into Word, Excel, PowerPoint and Outlook via add-ins; set as the default model in Grok Build.

VERIFIABLE FACTS

Every value stays attached to a source and date.

CAPABILITIES · Serving speed and token efficiencyDEVELOPER CLAIM

Served at 'fast-model' speeds of 80 TPS; claimed to deliver about twice the token efficiency versus the latest leading models on the same tasks.

WHAT CHANGED

Stored passport versions, without reconstructed history.

Passport updated16 facts
CONTEXT WINDOW · Context window+ 500,000 tokens
MODALITIES · Input modalities+ Text, images
MODALITIES · Output modalities+ Text
PRICING · Cached input price (≥200K input)+ $0.6 USD / 1M cached input tokens; prompt ≥200,000 tokens; all request tokens use this rate
PRICING · Cached input price (<200K input)+ $0.3 USD / 1M cached input tokens; prompt <200,000 tokens
PRICING · Input price (≥200K input)+ $4 USD / 1M input tokens; prompt ≥200,000 tokens; all request tokens use this rate
PRICING · Input price (<200K input)+ $2 USD / 1M input tokens; prompt <200,000 tokens
PRICING · Output price (≥200K input)+ $12 USD / 1M output tokens; prompt ≥200,000 tokens; all request tokens use this rate
PRICING · Output price (<200K input)+ $6 USD / 1M output tokens; prompt <200,000 tokens
Passport created7 facts