vs
2:11
overall capability wins (1 ties across 14 scored). Grok 4.6 leads overall.
- Reasoning1:4
- Coding0:5
- Arena1:0
- Agents0:1
- Tool use0:1
- Composite indices0:2
Verdict
Grok 4.6 leads coding (widest gap: +62.5 on WebDev Elo); Grok 4.5 ranks higher on LMArena (+13 Elo).
- Grok 4.6 leads coding (widest gap: +62.5 on WebDev Elo)
- Grok 4.5 ranks higher on LMArena (+13 Elo)
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Grok 4.5
- $3.50
- Grok 4.6
- $3.50
Price per 1M tokens
Full comparison
Organization
xAI
xAI
Family
Grok
Grok
License
Proprietary
Proprietary
Open weights
No
No
Release
Jul 9, 2026
Aug 12, 2026
Knowledge cutoff
-
2026-02-01
API / provider
xAI
xAI
Modalities
text, image → text
text, image → text
Specs
Context window
500K
500K
Max output
128K
-
Parameters
—
1.5T
Pricing
Input $/1M
$2
$2
Output $/1M
$6
$6
Blended $/1M (3∶1)
$3
$3
Speed
tok/s
100
-
TTFT (s)
0.35
-
Reasoning
MMLU-Pro
—
—
GPQA Diamond
93.1
94.9
Humanity's Last Exam
42.7
42.9
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
51.6
48.2
AA-LCR v1.1
79.3
80.3
CritPt
15.4
17.1
MMMU-Pro
80.4
—
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
81.6
88.4
Aider Polyglot
—
—
Terminal-Bench 3
15.7
26.5
BigCodeBench
—
—
SciCode
55.0
56.5
CursorBench
—
70.8
SWE-Rebench
63.8
—
NL2Repo-Bench
—
—
DeepSWE
—
67.0
WebDev Arena
1,555
1,618
Terminal-Bench 4.0
12.4
20.3
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,468
1,456
| Metric | Grok 4.5 | Grok 4.6 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | xAI | xAI | — |
| Family | Grok | Grok | — |
| License | Proprietary | Proprietary | — |
| Open weights | No | No | — |
| Release | Jul 9, 2026 | Aug 12, 2026 | — |
| Knowledge cutoff | - | 2026-02-01 | — |
| API / provider | xAI | xAI | — |
| Modalities | text, image → text | text, image → text | — |
| Specs | |||
| Context window | 500K | 500K | tie |
| Max output | 128K | - | — |
| Parameters | — | 1.5T | — |
| Pricing | |||
| Input $/1M | $2 | $2 | tie |
| Output $/1M | $6 | $6 | tie |
| Blended $/1M (3∶1) | $3 | $3 | tie |
| Speed | |||
| tok/s | 100 | - | — |
| TTFT (s) | 0.35 | - | — |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | 93.1 | 94.9 | B +1.8 pts |
| Humanity's Last Exam | 42.7 | 42.9 | B +0.2 pts |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 51.6 | 48.2 | A +3.4 pts |
| AA-LCR v1.1 | 79.3 | 80.3 | B +1.0 pts |
| CritPt | 15.4 | 17.1 | B +1.7 pts |
| MMMU-Pro | 80.4 | — | — |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | 81.6 | 88.4 | B +6.8 pts |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | 15.7 | 26.5 | B +10.8 pts |
| BigCodeBench | — | — | — |
| SciCode | 55.0 | 56.5 | B +1.5 pts |
| CursorBench | — | 70.8 | — |
| SWE-Rebench | 63.8 | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | 67.0 | — |
| WebDev Arena | 1,555 | 1,618 | B +63 Elo |
| Terminal-Bench 4.0 | 12.4 | 20.3 | B +7.9 pts |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,468 | 1,456 | A +13 Elo |
Grok 4.5
- Context
- 500K / 128K out
- Parameters
- —
- Price
- $2 / $6
- Speed
- 100 tok/s · 0.35s TTFT
- Modalities
- text, image → text
- License
- Proprietary
xAI/SpaceXAI public API flagship (July 2026) - Cursor-trained coding focus at $2/$6. For the Cursor IDE jointly-trained SKU and Fast tier, see Cursor Grok 4.5.
Grok 4.6
- Context
- 500K
- Parameters
- 1.5T
- Price
- $2 / $6
- Speed
- —
- Modalities
- text, image → text
- License
- Proprietary
xAI/SpaceXAI's August 2026 flagship (1.5T-parameter) tuned for long-running agents and ambitious interactive/visual work — 500k context at $2/$6, matches GPT-5.6 Sol on the Artificial Analysis Intelligence Index (61). A fast variant is available at 2x price.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.