Grok 4.6
Grok 4.6 is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.
Use it for
- Teams evaluating coding, rag, and agents
- Workloads that can use a 500k context window
- Buyers comparing 2 tracked provider routes
Do not use it for
- Workloads where another current model has stronger sourced task evidence
- Family
- Grok 4
- Released
- 2026-08-12
- Context
- 500k
- Knowledge cutoff
- 2026-02
- Specialization
- general
- Openness
- Proprietary
- License
- ProprietaryCommercial use: conditional
- Weights
- Not released
- Code
- Unknown
Cheapest of 2 routes · OpenRouter · cache read $0.500
About
Grok 4.6 is xAI/SpaceXAI's August 2026 frontier model for coding, agentic tasks, and knowledge work. It is publicly available as grok-4.6 on the xAI API and is also offered in Cursor and Grok Build. Official docs list text and image input, text output, a 500,000-token context window, no published text output limit, function calling, structured outputs, prompt caching, code execution, web search, X search, and configurable reasoning effort (low, medium, high default, or xhigh). Knowledge cutoff is February 1, 2026. Batch API is not supported.
Top use-case fit: coding, agents, and build tasks
Coding
Included by capability and metadata signals in the decision map.
RAG
Included by capability and metadata signals in the decision map.
Agents
Included by capability and metadata signals in the decision map.
Provider price ladder
Compare all 2Compare API pricing across 2 providers for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Cache | Route |
|---|---|---|---|---|
| OpenRouter | $2.00 | $6.00 | read $0.500 | Serverless |
| xAI Console | $2.00 | $6.00 | read $0.500 | Serverless |
Available via routers & gateways(1)
Capabilities
Benchmark peer barsfor Coding
No task-mapped benchmark peers are available for this model yet.
Benchmark scores(2)
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| DeepSWE 1.1 | 65.9 | DeepSWE v1.1, Grok 4.6 High (xAI first-party evals table)Observed 2026-08-12 | — | Source |
| CursorBench | 69.9 | CursorBench 3.2, Grok 4.6 High (xAI first-party evals table)Observed 2026-08-12 | Configuration: Grok 4.6 High Notes: xAI launch post evals table reports Grok 4.6 High at 69.9% on CursorBench v3.2. | Source |
Migration checks
No linked migration route is available for this model yet.
API versions
grok-4.6Cheapest of 2 routes · OpenRouter · cache read $0.500