Grok 4.5
xAIGrokProprietary
Description
Grok 4.5 is xAI's frontier model built for coding, agentic tasks, and knowledge work. Trained alongside Cursor across tens of thousands of NVIDIA GB300 GPUs with large-scale reinforcement learning focused on per-token intelligence, it features a 500K-token context window, configurable reasoning effort, native tool use, and strong token efficiency, and is served at fast-model speeds.
Release Date
2026-07-08
Parameters
—
Context Length
500K
Modalities
image, pdf, text
Capability Radar
100
general
20
coding
100
reasoning
77
scienceest.
80
agents
70
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agentic Capability | 55 | 56.0 | LS |
Benchmark Scores (LLM Stats)
Agents
GDPval-AA
1543.00 / 3000SR
Artificial Analysis Coding Agent Index v1.1
76.00 / 100SR
Terminal-Bench 2.1
83.3%SR
SWE-Bench Pro
64.7%SR
DeepSWE
53.0%SR
AutomationBench-AA
51.4%SR
Tau3 Banking
33.0%SR
SWE-Marathon
29.0%SR
Biology
GPQA
93.0%SR
General
Artificial Analysis
54.0%SR
Knowledge
AA-Omniscience Index
126.00 / 200SR
OmniScience
46.0%SR
AA Evaluation Indices
No AA evaluation data available
LLM Stats Category Scores
Legal100
Finance100
General100
Reasoning100
Agents100
Code20
Physics90
Biology90
Chemistry90
Tool Calling80
Pricing
Input Price$0 / 1M tokens
Output Price$0.00001 / 1M tokens
Blended Price (3:1)$0 / 1M tokens
Cache Read Price$0.5 / 1M tokens
Speed
No speed data available
Provider Price Ranking
Provider Price Ranking
3 providers
Cheapest: xAIMost Expensive: OpenCode Zen
ProviderInputOutput
1xAIPRIMARY
$0
$0.00001
2Abacus
$2
$6
3OpenCode Zen
$2
$6
Compare pricing across different API providers for this model.