Grok 4.5
xAIProprietaryReasoningVerifiedReleased Jul 8, 2026Input: text, image, file
Overall score
63.3
Rank 42 of 222 · 58.0–68.7
Context window
500K
500,000 tokens
Price per 1M tokens
$2.00 in · $6.00 out
✓ Verified on xAI's pricing page
Blended price
$3.00
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Multimodal82.9
- 2Knowledge#2877.0
- 3Agentic & tool use#3274.3
- 4Coding#3167.7
- 5Math62.9
- 6Reasoning#2754.1
Benchmark results (14)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Coding
| SWE-bench Pro | 64.7 | #13 of 69 |
| DeepSWE | 53.0 | #44 of 50 |
| PostTrainBench v1.1 | 23.4 | #13 of 15 |
| VulcanBench v3 | 89.9 | #1 of 14 |
Reasoning
| ARC-AGI-2 | 52.6 | #34 of 41 |
| ARC-AGI-1 | 85.67 | #30 of 36 |
| ARC-AGI-3 | 0.3 | #16 of 22 |
Agentic & tool use
| Terminal-Bench 2.1 | 83.3 | #21 of 68 |
| Terminal-Bench 3.0 | 15.7 | #7 of 13 |
Math
| FrontierMath v2 (Tiers 1-3) | 57.193 | #36 of 64 |
| FrontierMath v2 (Tier 4) | 24.390 | #35 of 51 |
Multilingual
| SWE Multilingual | 78.0 | #18 of 46 |
Multimodal
| Design Arena Website | 1281 | #23 of 91 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| ✓ Official (xAI) checked Oct 17, 2026 · grok-4.5 | $2.00 | $6.00 | — |
| BenchLM list price | $2.00 | $6.00 | $0.30 |
| OpenRouter x-ai/grok-4.5 | $2.00 | $6.00 | $0.30 |
Prompts under 200K tokens.
xAI's official July 2026 pricing page lists grok-4.5 at $2.00 input / $0.30 cached input / $6.00 output per million tokens below 200K tokens, with a 500K context window. Requests at or above 200K use the published $4.00 / $0.60 / $12.00 long-context tier; BenchLM stores the short-context rate.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text | 1448 | #56 · 39.9K votes |
| WebDev (coding) | 1553 | #43 · 11.1K votes |
| Vision | 1287 | #31 · 12.6K votes |
| Search | 1216 | #13 · 76.5K votes |
| Documents | 1452 | #26 · 5K votes |
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
Similar score
- DeepSeek V4 Pro 081363.2 · $0.99
- Qwen3.8-Flash-Next62.9 · —
- Gemini 3.6 Flash62.8 · $1.50
- Muse Spark Open62.8 · $2.00
- Qwen3.7 Max62.6 · $2.21