Oxyy is an independent API gateway, not affiliated with xAI.
Paid Grok text models run from $0.400 to $0.800 per 1M input tokens and $0.800 to $2.40 per 1M output tokens. Grok 4.6 costs $0.800 input / $2.40 output per 1M tokens, pay-as-you-go (standard price $2.00 / $6.00). The lowest-priced paid Grok text model is Grok Build 0.1 at $0.400 input / $0.800 output per 1M tokens.
Prices checked daily · last price change Sep 24, 2026 (discount changed: 70% → 60% off)
| Model | Context | Input | Output | Cached input | Saving | Released |
|---|---|---|---|---|---|---|
| Grok 4.6 (long prompts cost more, see note)grok-4.6 | 500K | Standard price $2.00, on Oxyy $0.800 | Standard price $6.00, on Oxyy $2.40 | $0.200 | 60% | Aug 12, 2026 |
| Grok 4.5 (long prompts cost more, see note)grok-4.5 | 500K | Standard price $2.00, on Oxyy $0.800 | Standard price $6.00, on Oxyy $2.40 | $0.120 | 60% | — |
| Grok 4.3 (long prompts cost more, see note)grok-4.3 | 1M | Standard price $1.25, on Oxyy $0.500 | Standard price $2.50, on Oxyy $1.00 | $0.080 | 60% | — |
| Grok 4.7grok-4.7 | 500K | Standard price $2.00, on Oxyy $0.800 | Standard price $6.00, on Oxyy $2.40 | $0.200 | 60% | Sep 21, 2026 |
| Grok Uncensored (long prompts cost more, see note)grok-uncensored | 500K | Standard price $2.00, on Oxyy $0.800 | Standard price $6.00, on Oxyy $2.40 | $0.120 | 60% | Jul 8, 2026 |
| Grok 4.20 (0309) Multi-Agent (long prompts cost more, see note)grok-4.20-multi-agent-0309 | 1M | Standard price $1.25, on Oxyy $0.500 | Standard price $2.50, on Oxyy $1.00 | $0.080 | 60% | — |
| Grok 4.20 (0309) Non-Reasoning (long prompts cost more, see note)grok-4.20-0309-non-reasoning | 1M | Standard price $1.25, on Oxyy $0.500 | Standard price $2.50, on Oxyy $1.00 | $0.080 | 60% | — |
| Grok 4.20 (0309) Reasoning (long prompts cost more, see note)grok-4.20-0309-reasoning | 1M | Standard price $1.25, on Oxyy $0.500 | Standard price $2.50, on Oxyy $1.00 | $0.080 | 60% | — |
| Grok Build 0.1 (long prompts cost more, see note)grok-build-0.1 | 256K | Standard price $1.00, on Oxyy $0.400 | Standard price $2.00, on Oxyy $0.800 | $0.080 | 60% | — |
| Model | Type | Price | Saving | Released |
|---|---|---|---|---|
| Grok Imagine Imagegrok-imagine-image | Image generation | Standard price $0.020, on Oxyy $0.008per image | 60% | — |
| Grok Imagine Videogrok-imagine-video | Text to video | From Standard price $0.050, on Oxyy $0.020per second of video | 60% | — |
| Grok Imagine Video 1.5grok-imagine-video-1.5 | Text to video | From Standard price $0.080, on Oxyy $0.032per second of video | 60% | — |
| Grok Voice API — Speech to Textgrok-stt | Speech to text | Standard price $0.100, on Oxyy $0.040per hour of audio | 60% | — |
| Grok Voice API — Text to Speechgrok-tts | Text to speech | Standard price $0.015, on Oxyy $0.006per 1K characters | 60% | — |
| Grok Voice Think Fast 2.0grok-voice-think-fast-2.0 | Realtime voice | Standard price $0.080, on Oxyy $0.032per minute of input | 60% | — |
| Grok Imagine Image 2.0grok-imagine-image-2.0 | Image generation | Not yet priced | — | — |
| Date | Model | Change |
|---|---|---|
| Sep 24, 2026 | 7 models | Discount changed: 70% → 60% off |
| Sep 22, 2026 | 7 models | Discount started: 70% off |
Estimate a month of Grok API usage: set your input and output volume and how much of your input is a repeated prompt prefix served from cache, or start from a preset.
Estimated monthly cost for 10M input and 2M output tokens, with 0% of input served from cache.
| Grok model | At standard price | On Oxyy | You save |
|---|---|---|---|
| Grok 4.6 (estimate is a lower bound, see note) | $32.00 | $12.80 | $19.20 (60%) |
| Grok 4.5 (estimate is a lower bound, see note) | $32.00 | $12.80 | $19.20 (60%) |
| Grok 4.3 (estimate is a lower bound, see note) | $17.50 | $7.00 | $10.50 (60%) |
| Grok 4.7 | $32.00 | $12.80 | $19.20 (60%) |
| Grok Uncensored (estimate is a lower bound, see note) | $32.00 | $12.80 | $19.20 (60%) |
| Grok 4.20 (0309) Multi-Agent (estimate is a lower bound, see note) | $17.50 | $7.00 | $10.50 (60%) |
| Grok 4.20 (0309) Non-Reasoning (estimate is a lower bound, see note) | $17.50 | $7.00 | $10.50 (60%) |
| Grok 4.20 (0309) Reasoning (estimate is a lower bound, see note) | $17.50 | $7.00 | $10.50 (60%) |
For each kind of work, the lowest-priced paid Grok model tagged for it and the one developers use most on Oxyy.
| Need | Lowest price | Most used on Oxyy |
|---|---|---|
| Coding | Grok Build 0.1$0.400 / $0.800 per 1M tokens | Grok 4.6 |
| Reasoning | Grok 4.3$0.500 / $1.00 per 1M tokens | Grok 4.6 |
| Agents and tool use | Grok Build 0.1$0.400 / $0.800 per 1M tokens | Grok 4.6 |
| Long documents | Grok 4.3$0.500 / $1.00 per 1M tokens | Grok 4.6 |
| Lowest cost | Grok Build 0.1$0.400 / $0.800 per 1M tokens | Grok 4.6 |
On Oxyy, Grok 4.6 costs $0.800 input / $2.40 output per 1M tokens at the pay-as-you-go price (standard price $2.00 / $6.00). Across paid Grok text models, input runs from $0.400 to $0.800 per 1M tokens and output from $0.800 to $2.40. Text to video starts at $0.020 per second of video, image generation starts at $0.008 per image and realtime voice starts at $0.032 per minute of input. There is no subscription: you pay only for what you use.
Grok Build 0.1 is the lowest-priced paid Grok text model on Oxyy, at $0.400 input / $0.800 output per 1M tokens.
Cached input on Grok 4.6 costs $0.200 per 1M tokens on Oxyy, against $0.800 for regular input. The price table on this page lists the cache rate of every model that has one.
Yes. Oxyy's API is OpenAI-compatible: set the base URL to https://api.oxyy.ai/v1, use your Oxyy API key and pass the model id, for example "grok-4.6", to /v1/chat/completions. Image models use /v1/images/generations, video models use /v1/videos/generations, transcription models use /v1/audio/transcriptions and text-to-speech models use /v1/audio/speech.
For some models, yes. Grok 4.6 costs $1.60 input / $4.80 output per 1M tokens once a prompt reaches 200K tokens, instead of $0.800 / $2.40; the whole request is priced at the higher rate.
No. Oxyy is an independent API gateway, not affiliated with or endorsed by xAI. It gives you one API key and one bill for models from many AI labs, xAI among them.
xAI's flagship model for code and everything else: agentic tool calling, minimal hallucinations, configurable reasoning.
Previous-generation Grok flagship. Same standard input and output rates as Grok 4.6 but a cheaper cached-input rate.
Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...
xAI frontier model for coding, knowledge work and STEM, co-trained with Cursor and tuned for agentic tool calling.
Pinned 0309 release of Grok 4.20 configured for multi-agent orchestration.
Pinned 0309 release of Grok 4.20 in non-reasoning mode, for low-latency responses.
Pinned 0309 release of Grok 4.20 in reasoning mode, with a 1M token context window.
Cost-efficient Grok model with a 1M token context window and a 20% Batch API discount.
xAI's coding-specialized model, the cheapest per token in the Grok lineup.
xAI's low-cost image generation and editing model, flat-priced across resolutions.
xAI's recommended image generation and editing model. Text and image in, image out.
Previous-generation, lower-cost video generation and editing model. Text, image and video in, video out.
xAI's recommended video generation model. Text, image and audio in, video out, with native audio.
xAI's speech-to-text transcription, available as batch REST or streaming.
xAI's text-to-speech mode, with custom voice support.
xAI's realtime speech-to-speech voice agent model.
