xAI model product
Grok 3 is an llm from xAI launched on February 17, 2025.
Updated Sep 8, 2026. Default version: Grok-3
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Grok-3.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAIME 2024 | Score93.30% | Rank03 | Participants53 | Percentile96.15% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score79.40% | Rank12 | Participants75 | Percentile85.14% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score78.00% | Rank18 | Participants64 | Percentile73.02% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score93.30% | Rank29 | Participants119 | Percentile76.27% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score84.60% | Rank64 | Participants247 | Percentile74.39% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,425.32 rating | Rank72 | Participants210 | Percentile66.03% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,411.47 rating | Rank98 | Participants210 | Percentile53.59% | EvidenceA | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryindustry medicine and healthcare | Rank28 | Rating / score1,466.98 | Votes1,606 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank42 | Rating / score1,461.08 | Votes392 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry entertainment and sports and media | Rank42 | Rating / score1,415.88 | Votes5,660 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank48 | Rating / score1,413.96 | Votes4,697 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank52 | Rating / score1,431.06 | Votes784 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank53 | Rating / score1,445.00 | Votes1,869 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank54 | Rating / score1,438.61 | Votes4,862 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank54 | Rating / score1,433.11 | Votes1,618 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank60 | Rating / score1,388.72 | Votes823 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry writing and literature and language | Rank63 | Rating / score1,410.69 | Votes7,725 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank69 | Rating / score1,374.09 | Votes520 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank70 | Rating / score1,418.43 | Votes22,978 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank70 | Rating / score1,418.65 | Votes784 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank71 | Rating / score1,437.67 | Votes17,053 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank71 | Rating / score1,441.38 | Votes5,818 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank72 | Rating / score1,425.32 | Votes32,415 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank72 | Rating / score1,448.67 | Votes392 | ObservationsN/A | Result date |
| Arenatext | Categoryinstruction following | Rank73 | Rating / score1,408.60 | Votes9,626 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank73 | Rating / score1,410.26 | Votes15,353 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank74 | Rating / score1,433.17 | Votes10,576 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank76 | Rating / score1,436.88 | Votes5,714 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank76 | Rating / score1,415.12 | Votes2,766 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank76 | Rating / score1,370.91 | Votes823 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank77 | Rating / score1,398.23 | Votes4,697 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry entertainment and sports and media | Rank77 | Rating / score1,397.15 | Votes5,660 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank77 | Rating / score1,405.79 | Votes7,725 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank77 | Rating / score1,414.22 | Votes1,618 | ObservationsN/A | Result date |
| Arenatext | Categorymulti turn | Rank78 | Rating / score1,424.58 | Votes4,648 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank80 | Rating / score1,446.00 | Votes1,606 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank83 | Rating / score1,437.86 | Votes9,098 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderHelicone | Provider model IDgrok-3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context131.1K | Updated |
| ProviderPoe | Provider model IDxai/grok-3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context131.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Grok-3. Runtime does not affect the capability score.
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderxAI | Output Speed100.00 tok/s | Catalog Latency0.70 s | Max Input128K | Max Output8K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGrok-3 | Released | LLMBoard53.30 | ParametersN/A | Context128K | Max output8K | Open weightsNo | LicenseProprietary |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Grok 3 and its available data.
Grok 3 is an llm from xAI that was trained with ten times more compute than Grok 2. The Grok 3 family includes Grok 3 Reasoning and Grok 3 Mini Reasoning for complex problem-solving.
Data as of 2026-09-08.
Common questions about Grok 3.
Grok 3's default version was released on Feb 17, 2025.
No official standard PAYG price is currently available for Grok 3. The lowest tracked third-party offer starts at $3 input and $15 output via Helicone.
Grok 3 was created by xAI.
The default version has a 128K token context window.
No. The default version is not marked as having publicly available weights.
2 provider offerings are linked to the default version.