Grok 4.6: xAI Ships a Long-Running Agent Model to OpenRouter
xAI announced Grok 4.6 on 12 August 2026, tuned for long-running agents and ambitious visual work. It is live on OpenRouter with a 500,000-token context at $2 per million input tokens, and the release date is confirmed by the official announcement.
AI Neural Narration
48kHz StudioFish Audio Neural Engine · Natural editorial narration
Key Takeaways
- check_circlexAI released Grok 4.6 on 12 August 2026, aimed at long-running agents, with reasoning on by default at high effort.
- check_circleOn OpenRouter it has a 500,000-token context and accepts text, images and files at $2 per million input and $6 per million output tokens.
- check_circleFine print: prompts over 200,000 tokens bill at $4 per million input and $12 per million output, and web search tool calls cost $5 each.
What Grok 4.6 changes
xAI announced Grok 4.6 on 12 August 2026. The model builds on Grok 4.5 with a particular focus on long-running agents and more ambitious interactive and visual work. The announcement says it stays with complex tasks across many steps, whether researching a topic, analysing information, working across a codebase, or turning an idea into a polished application.
On longer trajectories xAI reports more self-testing and verification, with the model checking its own work before moving on. It also produces stronger first passes on visual and interactive projects, which the announcement credits for faster iteration when starting from a concrete product idea.
Benchmarks
Grok 4.6 High posts an Artificial Analysis Intelligence Index score of 61, level with GPT-5.6 Sol Max, alongside 69.9% on CursorBench v3.2, 65.9% on DeepSWE v1.1, 1753 on GDPVal-AA v2 and 61.3% on FrontierCode v1.1 (Extended). Against Grok 4.5 High, the largest jumps are on agentic tasks: APEX-Agents moves from 47.1% to 57.5% and Terminal-Bench v3.0 from 15.7% to 26%.
Where it is available and what it costs
Grok 4.6 is live today in Cursor and Grok Build, with double included usage in both for the first week, plus the xAI API and partner platforms including OpenRouter, Vercel and Cloudflare. Pricing starts at $2 per million input tokens and $6 per million output tokens, with a fast variant at twice the price.
On OpenRouter the route is openrouter/x-ai/grok-4.6 with a 500,000-token context and text, image and file input. Reasoning is mandatory with high as the default effort, and xhigh, high, medium and low efforts are supported. Watch the fine print: OpenRouter bills prompts over 200,000 tokens at $4 per million input and $12 per million output, cached input reads at $0.50 per million, and web search tool calls at $5 each.
Detection in our index
AZ Labs endpoint monitoring first saw openrouter/x-ai/grok-4.6 at 15:45:17 UTC on 12 August. The OpenCode Zen route followed at 16:01:03 UTC, exactly 15 minutes and 46 seconds later.
Frequently Asked Questions
Is the Grok 4.6 release date confirmed?
Yes. xAI published the announcement on x.ai on 12 August 2026.
What does it cost on OpenRouter?
$2 per million input tokens and $6 per million output tokens. Prompts over 200,000 tokens bill at $4 per million input and $12 per million output, cached input reads at $0.50 per million, and web search tool calls cost $5 each.
Does it reason by default?
Yes. Reasoning is mandatory with high as the default effort. Xhigh, high, medium and low efforts are supported.
Where can I try it?
Cursor, Grok Build, the xAI API, and OpenRouter as openrouter/x-ai/grok-4.6. It is also on OpenCode Zen as grok-4.6.
Related Frontier Models & Releases
Explore verified specifications, benchmark results, and route pricing across alternative models in this class.
DeepSeek releases V4.1 Flash with native multimodal vision, 552B MoE and lower API rates
DeepSeek has launched DeepSeek-V4.1-Flash, a 552B-parameter mixture-of-experts model featuring a novel Causal Encoder–Decoder architecture with just 8B active input and 16B active output parameters. The release brings native vision, compresses KV cache storage by up to 8x, and slashes off-peak API pricing to $0.15 per million input tokens.
OpenAI Releases GPT-6 Astra: Next-Generation Flagship with 1.05M Context and Deep Multimodal Reasoning
OpenAI has officially launched GPT-6 Astra, its frontier flagship model featuring a 1,050,000-token context window, 128,000 max output tokens, and native tool-use for autonomous agent workflows.
Alibaba Qwen Releases Qwen3.8 Max: 2.4-Trillion Parameter Flagship with 1M Multimodal Context
Alibaba's Qwen team has launched Qwen3.8 Max, a 2.4T parameter Mixture-of-Experts model offering 1M token context, native video perception, and deep agent tool orchestration.