Grok 4.6 vs Grok 4.7
Too close to call on our weighted score (Grok 4.7 61, Grok 4.6 61). The right pick depends on what you value most.
Too close to call
It is close. Our weighted score puts them within a point (Grok 4.7 61/100, Grok 4.6 61/100), so choose by what matters most for your work: Grok 4.6 for raw capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond, FrontierMath Tiers 1–3, OTIS Mock AIME 2024–2025 and SimpleQA Verified), because Grok 4.7 has no Capabilities Index score yet.
- CapabilityGrok 4.6Shared benchmarks: Grok 4.6 77.1% · Grok 4.7 74.9%
- Lowest priceSame priceGrok 4.6 $3.00 · Grok 4.7 $3.00 per 1M tokens (3:1 blend)
- Longest contextAbout the sameGrok 4.6 500,000 · Grok 4.7 500,000 tokens
- Widest inputsGrok 4.7Grok 4.6: Text, Images · Grok 4.7: Text, Images, PDFs
- Self-hostingNo open weightsBoth are available only through APIs
| Measure | Weight | Grok 4.6 | Grok 4.7 |
|---|---|---|---|
| CapabilityShared benchmarks | 50% | 77 | 75 |
| Price | 25% | 27 | 27 |
| Inputs & features | 15% | 70 | 80 |
| Context window | 10% | 48 | 48 |
| Overall | 100% | 61/100 | 61/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification |  Grok 4.6xAI | Grok 4.7xAI |
|---|---|---|
| Capability | ||
| Capabilities Index (ECI) | 156.6 | — |
| ECI rank | #19 of 148 | — |
| GPQA DiamondGraduate-level science questions | 94.0% (best) | 92.7% |
| FrontierMath Tiers 1–3Research-level mathematics | 66.0% (best) | 53.0% |
| OTIS Mock AIME 2024–2025Competition mathematics | 99.2% (best) | 98.1% |
| SimpleQA VerifiedShort factual questions | 49.3% | 56.0% (best) |
| Price per million tokens | ||
| Input | $2.00 | $2.00 |
| Output | $6.00 | $6.00 |
| Cached input | $0.50 | $0.50 |
| Blended (3:1) | $3.00 | $3.00 |
| Long-context rate | Over 200K: $4.00 / $12.00 | Over 200K: $4.00 / $12.00 |
| Price source | Official xAI API | Official xAI API |
| Limits | ||
| Context window | 500,000 tokens | 500,000 tokens |
| Max output | 500,000 tokens | 500,000 tokens |
| Inputs and features | ||
| Text | Yes | Yes |
| Images | Yes | Yes |
| PDFs | No | Yes |
| Audio | No | No |
| Video | No | No |
| Reasoning | Yeslow · medium · high · xhigh | Yeslow · medium · high · xhigh |
| Tool calling | Yes | Yes |
| Structured output | Yes | Yes |
| Availability | ||
| Weights | Proprietary | Proprietary |
| API model ID | grok-4.6 | grok-4.7 |
| API providers | 29 (best) | 17 |
| Released | Aug 12, 2026 | Sep 21, 2026 |
| Knowledge cutoff | Feb 1, 2026 | May 2026 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Grok 4.6$32.00
- Grok 4.7$32.00
Which should you choose?
Which is better: Grok 4.6 or Grok 4.7?
It is close. Our weighted score puts them within a point (Grok 4.7 61/100, Grok 4.6 61/100), so choose by what matters most for your work: Grok 4.6 for raw capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond, FrontierMath Tiers 1–3, OTIS Mock AIME 2024–2025 and SimpleQA Verified), because Grok 4.7 has no Capabilities Index score yet.
Which is cheaper, Grok 4.6 or Grok 4.7?
Grok 4.6 and Grok 4.7 cost the same: $2.00 input / $6.00 output per million tokens.
Which scores higher on benchmarks?
Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond, FrontierMath Tiers 1–3, OTIS Mock AIME 2024–2025 and SimpleQA Verified): Grok 4.6 77.1% and Grok 4.7 74.9%. On individual benchmarks: GPQA Diamond — Grok 4.6 94.0%, Grok 4.7 92.7%; FrontierMath Tiers 1–3 — Grok 4.6 66.0%, Grok 4.7 53.0%; OTIS Mock AIME 2024–2025 — Grok 4.6 99.2%, Grok 4.7 98.1%; SimpleQA Verified — Grok 4.7 56.0%, Grok 4.6 49.3%.
Which is better for coding?
There are no published SWE-bench Verified results for Grok 4.6 and Grok 4.7 yet, so there is no like-for-like coding score. On overall capability, Grok 4.6 leads, which tends to carry over to coding, but test on your own codebase. Both support tool calling for agent workflows.
Which has the bigger context window?
Grok 4.6 and Grok 4.7 share the same 500,000-token context window. Maximum output per response: Grok 4.6 up to 500,000, Grok 4.7 up to 500,000 tokens.
Which can read images, PDFs, audio or video?
Grok 4.6 accepts text and images; Grok 4.7 accepts text, images and PDFs. Grok 4.7 handles the widest range of inputs.
Are any of these open source?
No. Grok 4.6 and Grok 4.7 are proprietary and only available through APIs and apps.
Which is newer?
Grok 4.7 is the newest, released Sep 21, 2026. Grok 4.6 came out Aug 12, 2026. Knowledge cutoff: Grok 4.6 Feb 1, 2026, Grok 4.7 May 2026.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
