What Happened to Grok Code Fast 1 and What Does It Redirect To?

Since the rollout of AI-powered coding assistants, the market has seen explosive interest fueled by innovations from companies like xAI and SpaceXAI, each pushing the boundaries of AI utility in software development. Among the buzzworthy tools, Grok Code Fast 1 once stood out as a promising solution for developers seeking rapid code generation and debugging capabilities. But as 2026 unfolds, users and observers alike are asking: What happened to Grok Code Fast 1, and where does it redirect now?

In this post, we examine the lifecycle of Grok Code Fast 1, its strategic redirection to Grok Build 0.1, current pricing models including a close look at SuperGrok Lite, the nuances grok 480p video limit of API per-token billing, and key changes shaping the consumer tiers vs the X-bundled offerings. We will also flag some critical "breakpoints" to watch out for in the near future.

Grok Code Fast 1 Retired: A Brief Timeline

Earlier in 2026, Grok, a division of xAI focused on AI-powered coding tools, announced the retirement of Grok Code Fast 1. This tool was popular among solo developers and small teams for its straightforward interface and relatively affordable pricing. However, evolution in AI capabilities and user demands necessitated a rewrite and rethink of the original product.

The official retirement date lined up closely with May 15, 2026 — an important breakpoint in Grok's product roadmap. After this date, attempts to access Grok Code Fast 1 services began redirecting users to Grok Build 0.1, the newer, more modular AI coding platform.

What Is Grok Build 0.1 and How Does It Differ?

Grok Build 0.1 marks a strategic shift from the one-size-fits-all approach of Grok Code Fast 1 toward a more customizable and granular AI coding assistant. This product focuses on providing flexible API access with transparent token-based billing and enhanced caching mechanisms that optimize cost and performance.

• Build 0.1 introduces multiple tiers — including $1 and $2 monthly plans — targeting micro-consumers and experimental users.
• It injects enhanced support for cached input tokens — effectively charging only for unique code inputs processed — which substantially lowers bills for repeated queries.
• API per-token billing is now the centerpiece of the pricing structure, emphasizing consumption-based costs rather than flat fees.

Compared to Grok Code Fast 1, which had more opaque usage limits and flat pricing slabs, Build 0.1 offers transparency and flexibility. However, this comes with a steeper learning curve and a need for users to monitor their token usage closely.

The 2026 Pricing Changes and Their Impact

Grok's migration from Code Fast 1 to Build 0.1 is accompanied by significant pricing shifts that align with industry trends set by competitors like SpaceXAI's AI platforms. Two key breakpoints define the pricing timeline:

1. May 15, 2026 — Official retirement of Grok Code Fast 1, redirecting users to Build 0.1.
2. August 12, 2026 — Implementation of new usage limits and updated free tier constraints.

The pricing structure for Build 0.1 is segmented to optimize for different user personas:

Tier Monthly Price Notes Build 0.1 Basic $1/month Entry-level, capped token usage Build 0.1 Plus $2/month Higher token limits, API access SuperGrok Lite $10/month or $100/year Effective $8.33/month with annual billing discount

Note: The annual billing discount math for SuperGrok Lite reflects a substantial 16.7% saving compared to monthly payments. It’s crucial to do this conversion rather than assume $10/month; annual pricing improves cost-efficiency.

Consumer Tiers vs. X-Bundled Tiers: What's the Difference?

Another ongoing theme in Grok’s product evolution is its differentiation between standard consumer tiers and X-bundled tiers. The X-bundles are specially packaged with xAI’s broader ecosystem — particularly integrating capabilities from SpaceXAI ventures — aiming to attract power users and enterprise customers.

• Consumer Tiers: Usually include Standalone Build 0.1 plans and SuperGrok Lite, targeting individuals and small teams. These have straightforward monthly or annual pricing and defined but somewhat restrictive free tier limits.
• X-Bundled Tiers: Integrated pricing with xAI main services, often with dynamic pricing adjustments based on multi-product usage, higher usage limits, and bundled token packages.

This segmentation addresses the vastly different consumption patterns — with X-bundled tiers offering more value to high-usage developers but often confusing casual users who must decipher bundle-based pricing.

Free Tier Limitations and Unknowns

One persistent source of dissatisfaction has been the evolution of free tier offerings following the retirement of Grok Code Fast 1. While Build 0.1 and SuperGrok Lite offer trial usage, limitations are increasingly stringent:

• Free tiers usually provide very limited token quotas, insufficient for heavy development work.
• There is ambiguity in how cached input tokens factor into free quota consumption — while repeated queries may be cheaper, it’s unclear if free tiers get the same benefit.
• Documentation is sparse on whether all free-tier activity counts equally toward token consumption or if some operations are excluded.

These gaps create hurdles for casual adopters looking to evaluate Build 0.1 before committing financially. Until xAI/Grok clarifies these points, users should assume free tiers are severely limited and plan budgets accordingly.

API Per-Token Billing and Cached Input Tokens Pricing

At the heart of Grok’s 2026 transition is a core shift to API per-token billing, a model now mainstream in B2B AI services. Unlike flat subscription plans of Grok Code Fast 1, this model charges users based on the precise *amount* of AI processing consumed (measured in tokens). Here are key implications:

• Transparency: Billing ties directly to usage, encouraging efficient coding queries.
• Cost Control: Developers can set caps or monitor token usage to stay within budgets.
• Cached Input Tokens: Queries with previously seen inputs incur lower or no additional cost, reducing expense for repeated code inspections.

However, this model demands careful usage tracking, which can be onerous for small shops without dedicated billing analysts. The effective per-token prices for Build 0.1 are still evolving, but early indications suggest a premium compared to legacy flat plans.

Summary: What You Need to Know

Here’s a concise recap for developers navigating this transition:

1. Grok Code Fast 1 is officially retired as of May 15, 2026, and its URLs now redirect to Grok Build 0.1.
2. Build 0.1 introduces flexible $1 and $2 monthly tiers, with usage-based token billing and caching strategies.
3. SuperGrok Lite remains a solid mid-tier option at $10/month or $100/year (effective $8.33/month), offering an annual discount worth calculating.
4. Consumer tiers are simpler but more limited, whereas X-bundled tiers combine Grok with xAI and SpaceXAI services and target power users.
5. Free tier offerings are restrictive and somewhat unclear in their exact usage limits and caching benefits.
6. API per-token billing represents the ongoing industry shift, requiring developers to adopt usage tracking and budgeting practices.
7. Watch the breakpoints May 15, 2026 and August 12, 2026 for further plan updates and limit changes.

Final Thoughts

The retirement of Grok Code Fast 1 is a textbook example of AI SaaS product evolution driven by scaling, precision pricing, and ecosystem integration. While Build 0.1’s per-token billing and caching mechanisms are forward-thinking, they impose new mental overhead on users accustomed to predictable flat-rate subscriptions.

For small teams or individual developers, weighing the benefits of SuperGrok Lite’s annual discount and clear pricing against the experimental Build 0.1 models is crucial. Meanwhile, keeping an eye on xAI and SpaceXAI’s AI innovations will illuminate where Grok’s roadmap lands next.

In an AI-driven world, adaptability is a mandate: knowing exactly what happened to Grok Code Fast 1, understanding its redirected home, and mastering the pricing logic can save time, money, and frustration.