SpaceXAI is the name xAI took after fully merging into SpaceX (the deal closed Feb 2, 2026). The Grok product name didn't change, and the API documentation still lives at docs.x.ai. The current text lineup is small: one flagship, one reasoning workhorse and one coding model (older Grok 4.6, 4.5 and 4.20 ids are still listed in the docs), plus Grok Imagine and Grok voice models.
Grok 4.7, the flagship, costs $2.00 input / $6.00 output per 1M tokens. Grok 4.3 is $1.25 / $2.50 with a 1M context, and Grok Build 0.1 for coding is $1.00 / $2.00. Once a prompt reaches 200K tokens, every model doubles its rates for the whole request.
| Model | Tier | Input / 1M | Output / 1M | Cache hit / 1M | Context | Notes |
|---|---|---|---|---|---|---|
| Grok 4.7 | Flagship | $2.00 | $6.00 | $0.50 | 500K | New flagship (Sep 21, 2026); larger base model, replaces 4.6 at the same price; ≥200K tokens: $4/$12, cache $1.00 |
| Grok 4.3 | Mid-tier | $1.25 | $2.50 | $0.20 | 1M | Reasoning workhorse; replaced Grok 4; ≥200K tokens: $2.50/$5, cache $0.40 |
| Grok Build 0.1 | Specialist | $1.00 | $2.00 | $0.20 | 256K | Code tasks; replaced Grok Code Fast 1; ≥200K tokens: $2/$4, cache $0.40 |
| Model | Support chatbot 10,000 conversations × (2,000 input + 400 output tokens) | Document Q&A 1,000 questions × (30,000 input + 800 output tokens) | Coding agent session 4M input tokens (90% cache hits) + 150,000 output tokens |
|---|---|---|---|
| Grok Build 0.1 | $28.00 | $31.60 | $1.42 |
| Grok 4.3 | $35.00 | $39.50 | $1.59 |
| Grok 4.7 | $64.00 | $64.80 | $3.50 |
Standard rates per 1M tokens from data on this site. The coding-agent row assumes 90% of input is read from the prompt cache (full input price where a model has no cache rate). Cache-write fees, batch discounts and long-context surcharges listed in the notes are not included. Updated November 2, 2026.
| Model | Type | Price | Notes |
|---|---|---|---|
| Grok Imagine 1.5 | Video | $0.080 / s | Cheapest frontier tier (480p–1080p) |
| Grok Imagine Video | Video | $0.050 / s | Older Grok Imagine video model (480p–720p); Grok Imagine 1.5 Lite is now cheaper |
| Grok Imagine 1.5 Lite | Video | $0.020 / s | Cheapest Grok Imagine video tier (grok-imagine-video-1.5-lite) |
| Grok Voice 3.0 | Voice | $0.08 / min | Real-time speech-to-speech (grok-voice-3.0); text input $0.004; replaces the deprecated grok-voice-2.0 at the same price |
| Model | Provider | Tier | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|---|
| Grok 4.7 | SpaceXAI | Flagship | $2.00 | $6.00 | 500K |
| Qwen3.8-Max | Alibaba | Flagship | $2.00 | $6.00 | 1M |
| Gemini 4 Argon | Flagship | $2.00 | $10.00 | 1M | |
| Kimi K3.1 Pro | Moonshot AI | Flagship | $2.00 | $10.00 | 1M |
| Gemini 3.1 Pro | Flagship | $2.00 | $12.00 | 1M | |
| Grok 4.3 | SpaceXAI | Mid-tier | $1.25 | $2.50 | 1M |
| Mistral Large 3 | Mistral | Mid-tier | $0.50 | $1.50 | 256K |
| DeepSeek V4-Pro | DeepSeek | Mid-tier | $0.66 | $1.98 | 1M |
| Mistral Large 4 | Mistral | Mid-tier | $0.70 | $2.10 | 1M |
| Gemini 3.8 Flash | Mid-tier | $0.75 | $3.75 | 1M |
SpaceXAI's cheapest model per tier (bold) against the four cheapest same-tier models from other providers, by output price.
SpaceXAI's docs say a request whose prompt reaches the long-context threshold is billed at the higher rate for all tokens in that request, not just the part above 200K. Grok 4.7 goes to $4/$12 per 1M, Grok 4.3 to $2.50/$5 and Grok Build 0.1 to $2/$4. A 210K-token prompt therefore costs about twice as much as a 190K one. If your documents sit near the line, trim or chunk them to stay below 200K. That's an easy way to halve the cost of those calls.
Grok 4.3 is cheaper on both input ($1.25 vs $2.00) and output ($2.50 vs $6.00). It also has the bigger window: 1M against Grok 4.7's 500K. Grok 4.7 is the newer flagship on a larger base model. Default to 4.3 for long-document reasoning and bulk work, and move a task to 4.7 when 4.3's answers aren't good enough. For code, Grok Build 0.1 is the cheapest of the three but has a 256K window.
Grok 4.7 has one of the lowest output prices among flagships: $6.00 per 1M against $10.00 for Gemini 4 Argon, $20.00 for Claude Opus 5.5 and $50.00 for GPT-6 Astra. That favours output-heavy jobs such as long reports, drafting and code generation. Its cache-hit rate of $0.50 is high compared with Argon's $0.10, though. In agent loops that re-read a large cached context on every step, Grok's advantage shrinks: in the coding-agent column above, cache reads make up about half of Grok 4.7's cost.
Both names refer to the same company. xAI rebranded as SpaceXAI after fully merging into SpaceX (deal closed Feb 2, 2026), and the Grok product name stayed the same. Model ids such as grok-4.7 and grok-4.3 are what you call in the API.
Only if you enable it. SpaceXAI's docs say real-time data needs the server-side search tools (Web Search or X Search) turned on in the request. Without them, Grok knows only what was in its training data.
SpaceXAI's model and pricing docs don't list free credits or a Batch discount; they note only that not every model accepts Batch API requests. Budget at the standard rates on this page. The free Grok chat plan is a consumer product and is separate from API billing.
At the flagship tier, Grok 4.7 ($2.00 / $6.00) is far cheaper than GPT-6 Astra ($10.00 / $50.00). In the mid tier, Grok 4.3 ($1.25 / $2.50) undercuts GPT-6.1 Sol ($2.00 / $10.00) on both input and output. GPT-6.1 Sol has cheaper cache hits ($0.10 vs $0.20), but even on our 90%-cached coding-agent example that doesn't make up for its higher input and output rates. Grok's 200K long-context surcharge kicks in earlier than OpenAI's 272K one, though.
How we choose: prices come from each vendor's official pricing page and are re-checked every week. We take no affiliate fees, so rankings aren't for sale.