Grok 4.5 (High)
SpaceXAIGrokProprietary
Description
Grok 4.5 is SpaceXAI's frontier model for coding, agentic tasks, and knowledge work. Trained alongside Cursor across tens of thousands of NVIDIA GB300 GPUs, it uses large-scale reinforcement learning focused on multi-step software engineering and other technical work. The model supports text and image inputs, reasoning, function calling, structured outputs, and a 500K-token context window, and is served at 80 tokens per second.
Release Date
2026-07-08
Parameters
—
Context Length
500K
Modalities
image, pdf, text
Capability Radar
40
general
70
coding
93
reasoning
70
science
40
agents
80
multimodal
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agentic Capability | 21 | 59.0 | LS |
| Code Ranking | 32 | 89.0 | AA |
| General Ranking | 66 | 66.0 | AA |
| Math Reasoning | 27 | 91.0 | LB |
| Reasoning | 25 | 87.0 | LB |
| Science | 25 | 82.0 | AA |
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))Agents
AutomationBench-AA
51.4%
Tau3 Banking
33.0%
Code
DeepSWE 1.0
62.0%SR
DeepSWE 1.1
54.0%
DeepSWE
53.0%SR
SWE-Marathon
29.0%SR
General
Artificial Analysis
54.0%
Knowledge
AA-Omniscience Index
126.00 / 200
OmniScience (non-hallucination rate)
46.0%
Reasoning
GPQANYU + Cohere + Anthropic (2023)
93.0%
Terminal-Bench 2.1
83.3%SR
SWE-Bench ProPrinceton NLP (2024)
64.7%SR
FrontierCode 1.1
42.4%
Terminal-Bench 4.0
12.4%
Science
OmniScience
52.0%
AA Evaluation Indices
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))93.1
Terminalbench V2 181.6
Lcr(Artificial Analysis)79.3
Coding Index(Artificial Analysis)72.4
Scicode(UIUC + Argonne National Lab (2024))55.0
Hle(Center for AI Safety + Scale AI (2025))42.7
Tau Banking42.1
Intelligence Index(Artificial Analysis)38.8
Terminalbench V4 010.6
LLM Stats Category Scores
(LLM Stats (zeroeval))Knowledge63
Science63
Reasoning16
General11
Physics90
Biology90
Chemistry90
Agents50
Code50
Tool Calling40
Pricing
Input Price$2 / 1M tokens
Output Price$6 / 1M tokens
Blended Price (3:1)$3 / 1M tokens
Cache Read Price$0.3 / 1M tokens
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
20 providers
Cheapest: xAIMost Expensive: Venice AI
ProviderInputOutput
1xAICheapest
$0
$0.00001
2SpaceXAIPRIMARY
$2
$6
3302.AI
$2
$6
4NanoGPT
$2
$6
5Abacus
$2
$6
6OpenRouter
$2
$6
7OpenCode Go
$2
$6
8ZenMux
$2
$6
9Kilo Gateway
$2
$6
10GitHub Copilot
$2
$6
11OpenCode Zen
$2
$6
12AIHubMix
$2
$6
13Requesty
$2
$6
14DevPass (LLM Gateway)
$2
$6
15CrossModel
$2
$6
16Pioneer
$2
$6
17DaoXE
$2
$6
18Ofox
$2
$6
19Opper
$2
$6
20Venice AI
$2.27
$6.8
Compare pricing across different API providers for this model.