If you searched for “Grok Code 1,” you most likely mean Grok Code Fast 1. xAI’s current model documentation identifies the API model as grok-build-0.1 and lists grok-code-fast-1 as an alias. For a new API integration, use the explicit current model name rather than relying on the older alias. (docs.x.ai)
What “Grok Code 1” refers to
xAI introduced Grok Code Fast 1 (grok-code-fast-1) as a model for agentic coding—workflows where a model reasons, calls tools, and edits code. “Grok Code 1” is not the name used in xAI’s announcement; it is best treated as shorthand, not a confirmed official product name. (docs.x.ai)
The current model documentation names the API model grok-build-0.1 and lists grok-code-fast-1 among its aliases. The model page describes it as intended for agentic software, engineering, and workflow tasks. (docs.x.ai)
Migration caveat: xAI’s migration guide contains conflicting directions for the retired grok-code-fast-1 slug: its replacement table maps it to grok-build-0.1, while its general routing text says retired slugs redirect to grok-4.3. Don’t assume the old alias selects the model or price you intend; specify grok-build-0.1 explicitly. (docs.x.ai)
Is it a good fit for coding work?
It may be worth testing for coding workflows where tool use and speed matter, but the available evidence does not establish that it is best for every coding task. xAI positions the model for agentic coding. That product positioning is not a guarantee of results on your codebase. (docs.x.ai)
A third-party Vals AI evaluation, dated August 27, 2025, reports these results for Grok Code Fast:
| Benchmark | Reported result | What the comparison suggests—and what it doesn’t |
|---|---|---|
| LCB | 62%; described as mid-pack | Vals AI called it comparable to Claude Sonnet 4 (Nonthinking) and said it cost a tenth as much in that comparison. The captured report gives no exact prices or billing unit. |
| IOI v1 | 4.3%; rank 8 of 12 | Vals AI reported Grok 4 at 26.2%, ranked first. This cautions against assuming the coding-focused model leads on every coding benchmark. |
| SWE-bench Verified | 57.6%; rank 4; 264.68 seconds | Vals AI reported Grok 4 at 58.6% and 704.8 seconds. Its report does not specify the latency method or test configuration. |
These are historical, attributed third-party results for Grok Code Fast, not current guarantees for Grok Build 0.1 or your workload. Use them as a reason to run your own evaluation, not as a production performance commitment. (docs.x.ai)
For a practical comparison, give the candidate model and your current choice the same representative bug fixes, feature changes, and multi-file tasks. Record test correctness, human correction needed, end-to-end time, and token cost. That will help you judge fit for your workload rather than relying on benchmark scores alone.
Production API: model ID, example, and cost
xAI’s announcement described Grok Build 0.1 as available through the xAI API in public beta. That is the status stated in the announcement, not confirmation of its current operational status or suitability for your production requirements. (docs.x.ai)
Here is a minimal illustrative request using the current documented model name and the official API overview’s Responses API request pattern. Confirm current access and account requirements before using it:
curl https://api.x.ai/v1/responses \
-H "Authorization: Bearer $XAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "grok-build-0.1",
"input": "Find the bug in this function and suggest a fix."
}'
The official API overview shows the Responses endpoint, bearer-token authentication, and a string input field. The model documentation lists grok-build-0.1 as the model name. (docs.x.ai)
API token pricing: xAI’s pricing documentation lists these Grok Build 0.1 rates in USD per million tokens:
| Pricing tier | Input | Cached input | Output |
|---|---|---|---|
| Short context | $1.00 | $0.20 | $2.00 |
| Long context | $2.00 | $0.40 | $4.00 |
The pricing table defines long context as 200,000 tokens or more. Once a request’s prompt reaches that threshold, the long-context rates apply to all tokens in the request; cached and uncached prompt tokens use their respective rates. (docs.x.ai)
Worked estimates, excluding any separate tool charges:
- Below the threshold: 100,000 uncached input tokens and 20,000 output tokens cost about $0.14:
$0.10input +$0.04output. - At the threshold: 200,000 uncached prompt tokens and 20,000 output tokens cost about $0.48:
$0.40input +$0.08output at the long-context rates.
These examples assume uncached input. The pricing page also states that requests sent to the US regional endpoint are billed at 1.1× the global token rates, a 10% premium. Check the rate applicable to your endpoint and include tool charges, if any, in your budget. (docs.x.ai)
Before choosing it
- Use the explicit model name: set
modeltogrok-build-0.1for a new integration; don’t rely on the retired alias without resolving the routing discrepancy. (docs.x.ai) - Confirm access and limits: check your account’s current access, rate limits, and the capabilities your application needs. The model page lists function calling and structured outputs; test the specific behavior and failure handling you depend on. (
docs.x.ai) - Estimate actual API cost: account for cached input, whether prompts reach the 200,000-token threshold, the endpoint’s applicable rates, and any tool charges. (
docs.x.ai) - Benchmark your own workload: compare correctness, review effort, latency, and cost against the model you already use. The Vals AI results are historical context, not a substitute for testing. (
docs.x.ai)
Bottom line: “Grok Code 1” most likely refers to Grok Code Fast 1; the current API model name is Grok Build 0.1 (grok-build-0.1). It is a candidate to evaluate for agentic coding, not an automatic production choice. Use the explicit model name, confirm current API access and limits, and test it on representative tasks.
Sources
Grok Build 0.1 — xAI model documentation
Grok Code Fast 1 — xAI model documentation
May 15, 2026 Model Retirement — xAI migration guide
Grok Code Fast 1 — xAI announcement
Grok Build 0.1 on API — xAI announcement
API Pricing — xAI documentation