xAI · released Aug 12, 2026
Grok 4.6
Grok 4.6 is xAI's current model and trails the leaders on most jobs.
3 months oldThinks before answering
- Input price
- $2 / M
- Output price
- $6 / M
- Context
- 500K tokens
- Longest reply
- 500K
- Reads
- text, image, pdf
- Knows up to
- 2026-02-01
Where it shines
- It scores 94 on GPQA.
- It costs $2 in, $6 out per million tokens.
Where it struggles
- Its text arena score of 1430 sits near the bottom of this lineup.
- Its 500,000-token context is half the size of most rivals.
Where you can use it
- xAI APIAPI$2 in / $6 out per million tokens.
- OpenRouterAPIListed as
x-ai/grok-4.6
How it scores on each job
Scores run 0 to 100 and blend every measure we have for that job. The number after the dot is its rank among the 33 models in the lineup.
The evidence
| Measure | Score | Rank in lineup | Source |
|---|---|---|---|
| Epoch Capabilities Index | 157 | #17 | Epoch AI |
| LMArena text | 1430 | #25 | LMArena |
| LMArena hard prompts | 1449 | #23 | LMArena |
| LMArena instruction following | 1432 | #22 | LMArena |
| LMArena multi-turn | 1426 | #25 | LMArena |
| LMArena creative writing | 1428 | #19 | LMArena |
| LMArena writing & language | 1431 | #22 | LMArena |
| LMArena coding | 1465 | #24 | LMArena |
| LMArena math | 1416 | #25 | LMArena |
| LMArena business & finance | 1417 | #25 | LMArena |
| LMArena legal & government | 1429 | #24 | LMArena |
| LMArena medicine | 1420 | #25 | LMArena |
| WebDev Arena | 1617 | #16 | LMArena |
| Vision Arena | 1263 | #18 | LMArena |
| Document Arena | 1452 | #7 | LMArena |
| Agent Arena | #26 | #16 | LMArena |
| GPQA Diamond | 94% | #10 | Epoch AI |
| FrontierMath | 66% | #21 | Epoch AI |
| SimpleQA Verified | 49% | #14 | Epoch AI |
| ARC-AGI-2 | 67% | #14 | Epoch AI |
Compare it with these
3 months old
Alibaba
Qwen3.8 Max
Qwen3.8 Max is Alibaba's flagship and gives solid all-round answers at a low output price.
Websites & appsProfessional work
- Price
- $2 in · $6 out
- Context
- 1M
Open weights3 months old
Z.ai
GLM-5.3
GLM-5.3 is Z.ai's open-weight model and handles chat decently at a low price.
Everyday questions
- Price
- $1.2 in · $4.18 out
- Context
- 1M
8 weeks old
Meta
Muse Spark 1.3
Muse Spark 1.3 is Meta's mid-price model and reads documents well for its cost.
Long documentsEveryday questions
- Price
- $1.25 in · $4.25 out
- Context
- 1M