xAI model product
Grok 4.6
6 is an xAI language model for coding, agentic tasks, and knowledge work.
Updated Sep 8, 2026. Default version: Grok 4.6
LLMBoard Score76.8Grok 4.6
Coverage0%8 benchmark families
Context window500KTokens
Official input price$2xAI API
Grok 4.6 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Grok 4.6 Benchmark Results
Benchmark scores for Grok 4.6.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAA-Briefcase | Score1,577.00 points | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAPEX-Agents | Score57.50% | Rank01 | Participants9 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAPEX-SWE | Score56.40% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkFrontierCode | Score61.30% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkHarvey LAB (Vals) | Score15.80% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkCursorBench v3.2 | Score69.90% | Rank02 | Participants2 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 3.0 | Score26.00% | Rank02 | Participants3 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkAA-Index | Score61.00% | Rank03 | Participants4 | Percentile33.33% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.78% | Rank03 | Participants49 | Percentile95.83% | EvidenceA | Evaluated |
| BenchmarkGDPval-AA | Score1,753.00 points | Rank05 | Participants9 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,624.86 rating | Rank08 | Participants97 | Percentile92.71% | EvidenceA | Evaluated |
| BenchmarkTerminal-Bench 4.0 | Score20.30% | Rank09 | Participants13 | Percentile33.33% | EvidenceB | Evaluated |
| BenchmarkDeepSWE 1.1 | Score65.90% | Rank12 | Participants31 | Percentile63.33% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score8.65% | Rank13 | Participants49 | Percentile75.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score7.83% | Rank14 | Participants49 | Percentile72.92% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score4.98% | Rank16 | Participants49 | Percentile68.75% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score3.86% | Rank16 | Participants49 | Percentile68.75% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score3.77% | Rank21 | Participants49 | Percentile58.33% | EvidenceA | Evaluated |
Grok 4.6 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent tool hallucination | Categoryoverall | Rank03 | Rating / score0.01 | VotesN/A | Observations1.8M | Result date |
| Arenawebdev | Categoryoverall | Rank08 | Rating / score1,624.86 | Votes3,487 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank08 | Rating / score1,624.86 | Votes3,487 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-react | Rank08 | Rating / score1,635.20 | Votes2,419 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank09 | Rating / score1,622.42 | Votes416 | ObservationsN/A | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank13 | Rating / score0.09 | VotesN/A | Observations32.5K | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank14 | Rating / score0.08 | VotesN/A | Observations12.9K | Result date |
| Arenaagent | Categoryoverall | Rank16 | Rating / score0.05 | VotesN/A | Observations1.9M | Result date |
| Arenaagent steerability | Categoryoverall | Rank16 | Rating / score0.04 | VotesN/A | Observations15K | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank21 | Rating / score0.04 | VotesN/A | Observations5.3K | Result date |
Grok 4.6 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $2 input, $6 output per 1M
- Official provider
- xAI
- Lowest third-party
- From $2 input, $6 output per 1M via OrcaRouter
- Tracked offerings
- 25
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderKenari | Provider model IDgrok-4-6 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context500K | Updated |
| ProviderOrcaRouter | Provider model IDgrok/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderNanoGPT | Provider model IDx-ai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderOpenRouter | Provider model IDx-ai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderCrossModel | Provider model IDx-ai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderOpenCode Go | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderOpper | Provider model IDxai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderxAI | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderMerge Gateway | Provider model IDxai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderCloudflare AI Gateway | Provider model IDxai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderPerplexity Agent | Provider model IDxai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderOfox | Provider model IDx-ai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderOpenCode Zen | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderRequesty | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderAIHubMix | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderVercel AI Gateway | Provider model IDspacexai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderEden AI | Provider model IDxai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgrok-4-6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderGitHub Copilot | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderKilo Gateway | Provider model IDx-ai/grok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderAbacus | Provider model IDgrok-4.6 | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context500K | Updated |
| ProviderAmazon Bedrock | Provider model IDxai.grok-4.6 | Regionglobal | Input / 1M$2.2 | Output / 1M$6.6 | Context500K | Updated |
| ProviderVenice AI | Provider model IDgrok-4-6 | Regionglobal | Input / 1M$2.27 | Output / 1M$6.8 | Context500K | Updated |
Grok 4.6 Runtime Performance
Provider-specific output speed and catalog latency for Grok 4.6. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderxAI | Output Speed132.69 tok/s | Catalog Latency1.70 s | Max Input500K | Max OutputN/A | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Grok 4.6 Specifications
Technical details for the model's default version.
- Version
- Grok 4.6
- Released
- Aug 12, 2026
- Knowledge cutoff
- Feb 1, 2026
- Parameters
- N/A
- Context window
- 500K
- Max output
- N/A
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Grok 4.6 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGrok 4.6 | Released | LLMBoard76.81 | ParametersN/A | Context500K | Max outputN/A | Open weightsNo | LicenseProprietary |
Models similar to Grok 4.6
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
XA
Grok 4.5
xAI
XA
Grok 4 Heavy
xAI
XA
Grok 4.20 Beta Reasoning
xAI
XA
Grok 4
xAI
XA
Grok 3
xAI
XA
Grok 4 Fast
xAI
What is Grok 4.6?
Key information about Grok 4.6 and its available data.
6 is a language model from xAI that supports text and image inputs, reasoning, function calling, structured outputs, and a 500K-token context window. It is designed for coding, agentic tasks, knowledge work, and interactive and visual work.
Data as of 2026-09-08.
FAQ
Common questions about Grok 4.6.
When was Grok 4.6 released?
Grok 4.6's default version was released on Aug 12, 2026.
How much does Grok 4.6 cost?
Grok 4.6's official API price is $2 per million input tokens and $6 per million output tokens via xAI. The lowest tracked third-party offer starts at $2 input and $6 output via OrcaRouter.
Who created Grok 4.6?
Grok 4.6 was created by xAI.
What is the context window for Grok 4.6?
The default version has a 500K token context window.
Is Grok 4.6 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Grok 4.6?
25 provider offerings are linked to the default version.