Grok 4.6
xAIProprietaryReleased Aug 12, 2026
On the arena.ai Agents leaderboard, Grok 4.6 has a task success rate of 0.74% (Xhigh, rank #27, 95% interval -0.37%–1.85%). Grok 4.6 ranks #70 on the LMArena Text leaderboard (style control) with an Arena score of 1455.4 (±4.7, 30,919 votes). Artificial Analysis Intelligence Index: 44.3. API price: $2.00 input / $6.00 output per 1M tokens. Context: 500K. Nearest rivals: Qwen3.7 Plus (1455.6) above and Hy3 (1453.4) below.
Quick answer
- Agents leaderboard
- #27 of 45 · 0.74% success
- Text (LMArena)
- #70 · 1455.4
- 17NAS overall
- #27 of 86
- API price (in / out)
- $2.00 / $6.00 per 1M tokens
- Closest rival
- Qwen3.7 Plus · 1455.6
- Cheaper alternative
- MiMo V2.5 Pro: Text 1466.4, output $0.87 (86% cheaper)
- Data updated
- Oct 22, 2026 · synced daily
Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter
Key metrics
Reasoning effort comparison
Artificial Analysis measurements per reasoning effort of the same model| Effort | Intelligence | Coding | tok/s | TTFT (s) | First answer (s) |
|---|---|---|---|---|---|
| highUsed in table | 44.3 | 76.8 | – | – | – |
| xhigh | 44.2 | 75.9 | – | – | – |
| medium | 42.8 | 74.4 | – | – | – |
| low | 35.1 | 66.3 | – | – | – |
Models ranked nearby
Sorted by Arena score| # | Model | Arena | Intel. | Output $ | tok/s |
|---|---|---|---|---|---|
| 48 |  GLM-5 | 1457.4 | 27.9 | $3.20 | – |
| 49 |  GPT-5.1 | 1455.8 | 24.7 | $10.00 | – |
| 50 |  Qwen3.7 Plus | 1455.6 | 25.2 | $1.60 | 55 |
| 51 | Grok 4.6This page | 1455.4 | 44.3 | $6.00 | – |
| 52 |  Hy3 | 1453.4 | 25.3 | $0.56 | 89 |
| 53 | GPT-5.6 Luna | 1452.5 | 37.3 | $1.20 | – |
| 54 |  Kimi K2.5 | 1449.9 | 23.5 | $3.00 | – |
More from xAI
Pricing
- Official API list price: $2.00 input and $6.00 output per 1M tokens (USD).
- At a typical 3:1 input-to-output ratio, the blended cost is about $3.00 per 1M tokens.
- Among 81 models with an output price, it is the #44 cheapest.
- The cheapest OpenRouter route charges $2.00 input / $6.00 output.
Sources & methodology
On the arena.ai Agents leaderboard, Grok 4.6 has a task success rate of 0.74% (Xhigh, rank #27, 95% interval -0.37%–1.85%).
- arena.ai Agents — agent task success rate, 95% interval, sessions and rank (via the public Hugging Face dataset, CC BY 4.0).
- LMArena — Arena score (Text leaderboard with style control, the same view arena.ai shows by default), 95% confidence interval, votes and rank.
- Artificial Analysis — Intelligence / Coding Index, output speed and latency (medians).
- OpenRouter — Route prices, context length and listing date.
Arena scores use the Bradley-Terry model (not Elo); ± is the half-width of the 95% confidence interval. Latency figures are Artificial Analysis medians; for reasoning models, time to first token and to first answer include the thinking phase. Data syncs daily at 14:20 (UTC+8).
Limits: Arena scores come from anonymous users voting on chats, so they reflect conversational preference rather than agent or coding benchmarks; with few votes the interval is wide and ranks move. Prices and speeds change as providers update them.
FAQ
How does Grok 4.6 do on the Agents leaderboard?
On the arena.ai Agents leaderboard, Grok 4.6 has a task success rate of 0.74% (Xhigh, rank #27, 95% interval -0.37%–1.85%).
What rank is Grok 4.6 right now?
Grok 4.6 is #51 on our leaderboard of 69 scored models (Arena score 1455.4); its raw LMArena rank is #70.
What does Grok 4.6 cost via API?
Input $2.00 / output $6.00 per 1M tokens (official API list price; the OpenRouter quote when no official price is available). At a typical 3:1 input-to-output ratio, the blended cost is about $3.00 per 1M tokens.
How close is Grok 4.6 to the models ranked next to it?
It is 0.2 points behind Qwen3.7 Plus (1455.6) and 2.0 points ahead of Hy3 (1453.4); gaps smaller than the confidence interval (±4.7) are not significant.
Is Grok 4.6 open source? Can it be self-hosted?
Grok 4.6 is a closed model, available only via official APIs or hosted services.
What is the context length of Grok 4.6?
500K (500,000 tokens, as listed on OpenRouter).
Grok 4.6 vs Qwen3.7 Plus: which is better?
On the LMArena Text leaderboard Grok 4.6 scores 1455.4 and Qwen3.7 Plus scores 1455.6 (a 0.2-point gap, within the confidence interval, so not significant). Agent task success: Grok 4.6 0.74%, Qwen3.7 Plus -6.69%. On price, Grok 4.6 costs $6.00 and Qwen3.7 Plus $1.60 per 1M output tokens.
Is there a cheaper model close to Grok 4.6?
Yes: MiMo V2.5 Pro scores 1466.4 on the Text leaderboard (Grok 4.6: 1455.4) and costs $0.87 per 1M output tokens, about 86% less than Grok 4.6 ($6.00).
Where does the Grok 4.6 data come from and how often is it updated?
Agent success rates come from the arena.ai Agents leaderboard (CC BY 4.0), Text scores from LMArena, intelligence and speed from Artificial Analysis, and prices and context from official list prices and OpenRouter. 17NAS syncs daily at 06:20 UTC; this page was last updated Oct 22, 2026.