Grok 4.6

Tool callingStructured outputReasoningImage input

Launch post Website Docs

xAI's frontier model for long-running agents, coding, knowledge work, and visual projects

Context: 500K
Released: 2mo ago
Input / 1M tokens: $2.00
Output / 1M tokens: $6.00
Cached input / 1M tokens: $0.50

Sources:models.devOpenRouter

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Other published benchmarks

Evaluation	Reported score and source
ARC-AGI-2	65.1% Third-party evaluator · grok-4.6_high Reasoning effort: high External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	27.6% Third-party evaluator · grok-4.6_low Reasoning effort: low External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	61.3% Third-party evaluator · grok-4.6_medium Reasoning effort: medium External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	67.1% Third-party evaluator · grok-4.6_xhigh Reasoning effort: xhigh External evaluation via Epoch AI Original-source terms apply
FrontierMath-Tier-4-v2-Private	31.7% Epoch AI · grok-4.6_xhigh Reasoning effort: xhigh Epoch AI evaluation
FrontierMath-Tiers-1-3-v2-Private	66.0% Epoch AI · grok-4.6_xhigh Reasoning effort: xhigh Epoch AI evaluation
GPQA diamond	94.0% Epoch AI · grok-4.6_high Reasoning effort: high Epoch AI evaluation
GPQA diamond	93.2% Epoch AI · grok-4.6_xhigh Reasoning effort: xhigh Epoch AI evaluation
OTIS Mock AIME 2024-2025	97.8% Epoch AI · grok-4.6_high Reasoning effort: high Epoch AI evaluation
OTIS Mock AIME 2024-2025	99.2% Epoch AI · grok-4.6_xhigh Reasoning effort: xhigh Epoch AI evaluation
SimpleQA Verified	49.3% Epoch AI · grok-4.6_high Reasoning effort: high Epoch AI evaluation
SimpleQA Verified	48.9% Epoch AI · grok-4.6_xhigh Reasoning effort: xhigh Epoch AI evaluation

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Specifications

Developer: xAI
Model family: Grok
Type: Language model
Predecessor: —
Successor: —
Released: August 12, 2026
Model ID: x-ai/grok-4.6
Openness: Closed
License: —
Input modalities: text, image, file
Output modalities: text
Modality: text+image+file->text
Tokenizer: Grok
Max output tokens: 450K
Knowledge cutoff: 2026-02-01
Hugging Face: —
Source freshness: Metadata synced Oct 8, 2026, 10:17 PM UTC

Additional source evidence

Source model: xai/grok-4.6
Source family: grok
Last updated: 2026-08-12
Knowledge cutoff: 2026-02-01
Model context: 500K
Model output limit: 500K
Open weights: No
File attachments: Yes
Reasoning: Yes
Tool calling: Yes
Structured output: Yes
Temperature control: Yes

Published benchmarks

Benchmark	Score	Evaluation details	Evidence
CursorBench	40.4 · score	high effort · 4.0 · 2026-09-21	Source
DeepSWE	65.2 · score	high effort · 1.1 · 2026-09-21	Source
EEBench	53 · score	high effort · 2026-09-21	Source
AA-Briefcase	1546 · Elo	high effort · 1.1 · 2026-09-21	Source
Terminal-Bench	20.3 · score	high effort · 4.0 · 2026-09-21	Source
Harvey Legal Agent Benchmark	15.8 · score	high effort · 2026-09-21	Source
HealthBench	48.5 · score	high effort; professional · 2026-09-21	Source
GDPval	1605 · Elo	high effort · 2026-09-21	Source

Complete source record (JSON)

{
  "id": "xai/grok-4.6",
  "name": "Grok 4.6",
  "description": "xAI's frontier model for long-running agents, coding, knowledge work, and visual projects",
  "family": "grok",
  "last_updated": "2026-08-12",
  "knowledge": "2026-02-01",
  "attachment": true,
  "reasoning": true,
  "tool_call": true,
  "structured_output": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text",
      "image"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 500000,
    "output": 500000
  },
  "benchmarks": [
    {
      "name": "CursorBench",
      "score": 40.4,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "score",
      "variant": "high effort",
      "version": "4.0",
      "date": "2026-09-21"
    },
    {
      "name": "DeepSWE",
      "score": 65.2,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "score",
      "variant": "high effort",
      "version": "1.1",
      "date": "2026-09-21"
    },
    {
      "name": "EEBench",
      "score": 53,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "score",
      "variant": "high effort",
      "date": "2026-09-21"
    },
    {
      "name": "AA-Briefcase",
      "score": 1546,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "Elo",
      "variant": "high effort",
      "version": "1.1",
      "date": "2026-09-21"
    },
    {
      "name": "Terminal-Bench",
      "score": 20.3,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "score",
      "variant": "high effort",
      "version": "4.0",
      "date": "2026-09-21"
    },
    {
      "name": "Harvey Legal Agent Benchmark",
      "score": 15.8,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "score",
      "variant": "high effort",
      "date": "2026-09-21"
    },
    {
      "name": "HealthBench",
      "score": 48.5,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "score",
      "variant": "high effort; professional",
      "date": "2026-09-21"
    },
    {
      "name": "GDPval",
      "score": 1605,
      "source": "https://x.ai/news/grok-4-7",
      "metric": "Elo",
      "variant": "high effort",
      "date": "2026-09-21"
    }
  ],
  "release_date": "2026-08-12",
  "open_weights": false
}

Infrastructure

Provider availability

All providers

Provider	Endpoint	Context	Input /M tokens	Output /M tokens	Cached input /M tokens	Cache write /M tokens	Uptime 1d	Data handling
xAI	xai/zdr	500K	$2.00	$6.00	$0.50	—	100.00%	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
xAI	xai	500K	$2.00	$6.00	$0.50	—	100.00%	ZDR: No Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
Amazon Bedrock	amazon-bedrock/us-west-2	500K	$2.20	$6.60	$0.55	$0.00	—	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
xAI	xai/zdr/us	500K	$2.20	$6.60	$0.55	—	100.00%	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
xAI	xai/zdr/priority	500K	$4.00	$12.00	$1.00	—	—	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15
xAI	xai/priority	500K	$4.00	$12.00	$1.00	—	—	ZDR: No Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15

Provider data:OpenRouter