Free Grok 4.5 trial for Felo Pro users

Try Grok 4.5 Freein Felo Search

Felo Search now supports SpaceXAI’s Grok 4.5 model. Felo Pro users can try Grok 4.5 free online during the 7-day launch trial and test coding, source-backed research, and agentic search with live web context.

Felo Pro subscribers can try Grok 4.5 free for 7 days, from July 9 to July 16, 2026. The free trial campaign ends automatically after the trial window.
Grok 4.5 on Felo Search
$2 / $6
Official API price
7 days
7-day free trial
Jul 8
2026 release
Grok 4.5 was announced by SpaceXAI one day before this Felo trial page.
$2 / $6
Official API price
SpaceXAI listed price per 1M input/output tokens. This is not Felo subscription pricing.
~80 TPS
Official speed claim
SpaceXAI reports the model is served around 80 tokens per second.
7 days
7-day free trial
Try Grok 4.5 free online from July 9 to July 16, 2026.

Free Grok 4.5 trial for Felo Pro

This free Grok 4.5 trial is a Felo launch campaign, separate from SpaceXAI’s API pricing. It starts on July 9, 2026 and ends automatically on July 16, 2026.

Source: Felo campaign terms for this launch page

01

Open Felo Search

The CTA opens Felo Search with search_model=grok-4.5 selected.

02

Use a Felo Pro account

Eligible Pro subscribers can try Grok 4.5 free during the 7-day launch window.

03

Bring real work

Test coding research, model comparison, market briefs, and source-backed questions instead of generic prompts.

04

Access ends automatically

The free trial window ends on July 16, 2026 without a separate cancellation step.

Official Grok 4.5 data, with sources marked

This free Grok 4.5 trial page uses Grok launch data, but labels who made each claim. Official model pricing is shown as SpaceXAI API pricing, not as a Felo plan price.

Built for coding and agents

Built for coding and agents

SpaceXAI describes Grok 4.5 as built for coding, agentic tasks, and knowledge work. Cursor says the model uses developer interaction data in its training mix.

Source: SpaceXAI and Cursor official launch posts

Price-performance angle

Price-performance angle

SpaceXAI lists standard Grok 4.5 at $2 per 1M input tokens and $6 per 1M output tokens. We present this as official model API pricing only.

Source: SpaceXAI official model pricing

Long-context work

Long-context work

xAI documentation lists a 500,000-token context window. In Felo Search, that makes it worth testing on source-heavy research and long comparison threads.

Source: xAI developer documentation

Efficiency claim needs attribution

Efficiency claim needs attribution

SpaceXAI reports up to 4.2x fewer output tokens than Claude Opus 4.8 on SWE-bench Pro. Because this is self-reported, the page labels it as a SpaceXAI claim.

Source: SpaceXAI official benchmark notes

Official SpaceXAI API pricing

These prices are published by SpaceXAI for model API usage. They are not Felo subscription prices and may change on the provider side.

Source: SpaceXAI official model pricing, checked July 9, 2026

Model
Input
Output
Note
Grok 4.5
$2 / 1M tokens
$6 / 1M tokens
Standard model price listed by SpaceXAI
Cached input
$0.50 / 1M tokens
-
Provider-side cached input pricing where available

Felo’s 7-day free Grok 4.5 Pro trial is a launch campaign for Felo Search access. The table above is official SpaceXAI API pricing, not a Felo billing table.

Benchmarks

Use official numbers, but do not overclaim

Grok 4.5 has strong launch data, but not every score is independently verified. The safest framing is price-performance, coding-agent fit, and real-task evaluation inside Felo Search.

Claim
Published data
Source
How this page frames it
Serving speed
Around 80 tokens per second
SpaceXAI official
Shown as an official provider claim
Context window
500,000 tokens
xAI developer docs
Useful for long research, with provider billing caveats
SWE-bench Pro
64.7% and 4.2x fewer output tokens vs Opus 4.8
SpaceXAI official
Self-reported; use with attribution
Terminal-Bench 2.1
83.3%
SpaceXAI official
Strong coding-agent signal, still provider-reported
Professional tasks
29% GDPval+ pass rate in one independent evaluation
Snorkel AI
Useful third-party signal for knowledge work

What to test during the free Grok 4.5 trial

Use the free trial window for prompts where source quality, answer speed, and model choice matter. Felo Search lets you evaluate Grok 4.5 free with current web context.

Source-backed research

Source-backed research

Ask a current question and check whether Grok 4.5 keeps evidence, assumptions, and recommendations separate.

Coding diagnosis

Coding diagnosis

Paste an error, migration problem, or architecture tradeoff. Look for a repair path you can verify.

Longer threads

Longer threads

Run follow-up questions and see if the model keeps decisions consistent across a research session.

Model comparison

Model comparison

Compare Grok 4.5 with GPT, Claude, Gemini, and other Felo-supported models on the same task.

Cost-aware tasks

Cost-aware tasks

For repeated workflows, evaluate cost per useful answer, not just a single benchmark headline.

Agentic planning

Agentic planning

Use it for task decomposition, source checklists, implementation plans, and multi-step decisions.

Early X feedback: practical, not blind hype

Felo X Search found 225 unique Grok 4.5 posts on July 9, 2026. Users repeatedly discussed price, Cursor availability, speed, and token efficiency, while cautioning that it does not win every benchmark.

Price is the headline

Many users called out the $2/$6 official API price and framed Grok 4.5 as a monthly-bill win for frequent model use.

Cursor distribution matters

Developers noticed that Grok 4.5 is available where they already work, especially Cursor and Grok Build.

Benchmark claims need context

Critical posts argued that official charts should be read carefully and that independent evaluations matter.

Best free Grok 4.5 trial use cases

Start the free trial where current information and repeated model calls matter. Grok 4.5 is especially interesting when speed and official API price affect workflow choice.

Coding research

Coding research

Compare framework changes, migration paths, API behavior, and debugging ideas with current sources attached.

Agentic search planning

Agentic search planning

Turn an open-ended task into a plan, source checklist, and next action list.

Market and product briefs

Market and product briefs

Summarize competing claims, compare evidence, and draft decision memos with citations.

Legal and policy scanning

Legal and policy scanning

Use Grok 4.5 for first-pass document summaries, then verify important details with source links.

Model evaluation

Model evaluation

Run the same prompt across Felo-supported models and compare answer shape, source use, and speed.

High-volume workflows

High-volume workflows

If your team repeats similar research prompts, evaluate Grok 4.5 for cost per useful answer.

Two ways to try Grok 4.5 free on Felo

Felo AI Search

Open Felo Search with Grok 4.5 selected. This is the fastest path to try Grok 4.5 free with live web context and citations.

Open Free Grok 4.5 Search

Felo LLM Playground

Use the Felo LLM Playground to compare Grok 4.5 with other models on the same prompt before choosing a default workflow.

Open Playground

Frequently Asked Questions

Grok 4.5 is SpaceXAI's model released on July 8, 2026. SpaceXAI positions it for coding, agentic tasks, and knowledge work, with official materials highlighting speed, context length, and price-performance.

Try Grok 4.5 Free on Felo Search

Use the Felo Pro 7-day free trial window to test Grok 4.5 with coding research, source-backed search, and model comparisons.

Try Grok 4.5 Free

Opens Felo Search with Grok 4.5 selected where available for the free Grok 4.5 trial

Official Grok 4.5 model data and API pricing are attributed to SpaceXAI/xAI official materials. Cursor availability is attributed to Cursor official materials. X user feedback is summarized from Felo X Search results collected on July 9, 2026.