Grok 4.6 (high)
AI model by xAI. Real-time pricing and benchmark data.
Benchmarks
Editorial Analysis
Profile. Grok 4.6 (high) is positioned in the mid-tier at $6.00/M output tokens. The benchmark profile is balanced — strongest on Coding Index (76.8) — making it a default choice for teams that need general-purpose reasoning without paying frontier-model prices.
Latency. On latency, Grok 4.6 (high) reports 72 tok/s output speed and 36220ms time-to-first-token. Output speed is on the slower end; for real-time chat consider pairing with a streaming front-end or routing latency-sensitive requests to a faster model.
Cost model. For a workload of 1M input tokens + 100K output tokens, Grok 4.6 (high) costs $2.60 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.
Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-08-15.
Compare with similar models
| Model | Input | Output | Speed |
|---|---|---|---|
Grok 4.6 (high)Current | $2.00 | $6.00 | 72 tok/s |
| Mistral Large 2 (Jul '24) | $2.00 | $6.00 | — |
| Qwen3 Max (Preview) | $1.20 | $6.00 | 63 tok/s |
| Qwen3.8 Max | $2.00 | $6.00 | 47 tok/s |
| Qwen3 Max | $1.20 | $6.00 | 69 tok/s |
| Qwen3.8 2.4T A95B | $2.00 | $6.00 | 49 tok/s |