Grok 4.6
xAIProprietaryReasoningVerifiedReleased Aug 12, 2026Input: text, image, file
Overall score
67.1
Rank 28 of 222 · 63.8–70.4
Context window
500K
500,000 tokens
Price per 1M tokens
$2.00 in · $6.00 out
✓ Verified on xAI's pricing page
Blended price
$3.00
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Agentic & tool use#977.6
- 2Coding#2477.3
- 3Knowledge#2075.9
- 4Math70.1
- 5Reasoning#2261.4
Benchmark results (19)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| MazeBench | 1 | #13 of 42 |
| ApprenticeBench | 13 | #13 of 18 |
| Terminal-Bench 3.0 | 26.5 | #4 of 13 |
| CWE-bench v1 | 57.0 | #6 of 11 |
| APEX-Agents | 57.5 | #1 of 8 |
Coding
| DeepSWE | 65.9 | #29 of 50 |
| Bug Hunt Bench | 27.0 | #13 of 23 |
| FrontierSWE v2 | 25.3 | #14 of 20 |
| VulcanBench v3 | 87.0 | #4 of 14 |
| FrontierCode 1.1 Extended | 61.3 | #4 of 8 |
Reasoning
| ARC-AGI-2 | 67.1 | #22 of 41 |
| ARC-AGI-1 | 87.00 | #26 of 36 |
| ARC-AGI-3 | 2.1 | #8 of 22 |
Math
| FrontierMath v2 (Tiers 1-3) | 65.965 | #28 of 64 |
| FrontierMath v2 (Tier 4) | 31.707 | #24 of 51 |
Multimodal
| Design Arena Website | 1289 | #20 of 91 |
Instruction following
| Gray Swan IPI (15 attempts)(lower is better) | 51.8 | #11 of 12 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| ✓ Official (xAI) checked Oct 17, 2026 · grok-4.6 | $2.00 | $6.00 | — |
| BenchLM list price | $2.00 | $6.00 | $0.50 |
| OpenRouter x-ai/grok-4.6 | $2.00 | $6.00 | $0.50 |
Prompts under 200K tokens.
xAI's August 12, 2026 release notes list grok-4.6 at $2.00 input / $0.50 cached input / $6.00 output per million tokens below 200K prompt tokens. Prompts above 200K use the published $4.00 / $1.00 / $12.00 long-context tier; BenchLM stores the short-context rate.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text(high) | 1428 | #99 · 29K votes |
| WebDev (coding)(High) | 1617 | #27 · 10.5K votes |
| Vision(High) | 1263 | #52 · 7.3K votes |
| Documents(high) | 1452 | #25 · 2.3K votes |
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
Similar score
- MiniMax M3.1 Flash67.1 · $0.26
- Claude Haiku 5.567.2 · $0.20
- Gemini 3.7 Flash66.9 · $1.50
- GPT-5.466.7 · $5.63
- GPT-5.2 Pro66.7 · $57.75