xAI model product
Grok 4.3
3 is a general-purpose large language model from xAI with a 1 million token context window and support for reasoning, function calling, and structured outputs.
Updated Sep 8, 2026. Default version: Grok 4.3
LLMBoard Score39.2Grok 4.3
Coverage0%1 benchmark families
Context window1MTokens
Official input price$1.25xAI API
Grok 4.3 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Grok 4.3 Benchmark Results
Benchmark scores for Grok 4.3.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkLM Arena Search Style Control | Score1,193.23 rating | Rank14 | Participants28 | Percentile51.85% | EvidenceA | Evaluated |
| BenchmarkLM Arena Search Factuality | Score1,173.56 rating | Rank18 | Participants28 | Percentile37.04% | EvidenceA | Evaluated |
| BenchmarkFinance Agent v2 | Score37.71% | Rank19 | Participants26 | Percentile28.00% | EvidenceB | Evaluated |
| BenchmarkLM Arena Search | Score1,164.80 rating | Rank19 | Participants28 | Percentile33.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.67% | Rank27 | Participants49 | Percentile45.83% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,241.34 rating | Rank40 | Participants107 | Percentile63.21% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score-11.57% | Rank44 | Participants49 | Percentile10.42% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score-13.46% | Rank44 | Participants49 | Percentile10.42% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score-13.13% | Rank44 | Participants49 | Percentile10.42% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score-19.71% | Rank45 | Participants49 | Percentile8.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score-11.11% | Rank49 | Participants49 | Percentile0.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision | Score1,227.77 rating | Rank54 | Participants107 | Percentile50.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,442.96 rating | Rank61 | Participants210 | Percentile71.29% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,356.49 rating | Rank75 | Participants97 | Percentile22.92% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,429.12 rating | Rank80 | Participants121 | Percentile34.17% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,398.36 rating | Rank104 | Participants210 | Percentile50.72% | EvidenceA | Evaluated |
Grok 4.3 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenasearch style control | Categoryoverall | Rank14 | Rating / score1,193.23 | Votes91,483 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank16 | Rating / score1,224.16 | Votes102 | ObservationsN/A | Result date |
| Arenasearch factuality | Categoryoverall | Rank18 | Rating / score1,173.56 | Votes91,483 | ObservationsN/A | Result date |
| Arenasearch | Categoryoverall | Rank19 | Rating / score1,164.80 | Votes91,483 | ObservationsN/A | Result date |
| Arenavision | Categoryentity recognition | Rank21 | Rating / score1,231.01 | Votes102 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryjapanese | Rank22 | Rating / score1,411.60 | Votes486 | ObservationsN/A | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank27 | Rating / score0.01 | VotesN/A | Observations1.1M | Result date |
| Arenavision style control | Categorycreative writing vision | Rank27 | Rating / score1,251.65 | Votes1,331 | ObservationsN/A | Result date |
| Arenawebdev | Categoryimage to webdev | Rank33 | Rating / score1,372.71 | Votes5,083 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank35 | Rating / score1,389.43 | Votes487 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank35 | Rating / score1,235.59 | Votes742 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank36 | Rating / score1,428.13 | Votes763 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank36 | Rating / score1,284.06 | Votes1,065 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank37 | Rating / score1,246.75 | Votes8,652 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank38 | Rating / score1,364.66 | Votes583 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank39 | Rating / score1,395.20 | Votes637 | ObservationsN/A | Result date |
| Arenavision style control | Categorydiagram | Rank40 | Rating / score1,265.09 | Votes5,544 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank40 | Rating / score1,241.34 | Votes20,787 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing vision | Rank41 | Rating / score1,240.70 | Votes1,331 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank43 | Rating / score1,229.57 | Votes742 | ObservationsN/A | Result date |
| Arenavision style control | Categoryocr | Rank43 | Rating / score1,249.34 | Votes14,829 | ObservationsN/A | Result date |
| Arenaagent | Categoryoverall | Rank44 | Rating / score-0.12 | VotesN/A | Observations1.6M | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank44 | Rating / score-0.13 | VotesN/A | Observations16.3K | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank44 | Rating / score-0.13 | VotesN/A | Observations57.4K | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank45 | Rating / score-0.20 | VotesN/A | Observations49.6K | Result date |
| Arenatext style control | Categorykorean | Rank47 | Rating / score1,404.21 | Votes1,041 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhomework | Rank47 | Rating / score1,252.80 | Votes2,674 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank48 | Rating / score1,446.72 | Votes1,124 | ObservationsN/A | Result date |
| Arenaagent steerability | Categoryoverall | Rank49 | Rating / score-0.11 | VotesN/A | Observations74.9K | Result date |
| Arenavision | Categorychinese | Rank49 | Rating / score1,263.61 | Votes1,065 | ObservationsN/A | Result date |
Grok 4.3 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $1.25 input, $2.5 output per 1M
- Official provider
- xAI
- Lowest third-party
- From $1.25 input, $2.5 output per 1M via OrcaRouter
- Tracked offerings
- 25
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderOrcaRouter | Provider model IDgrok/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderImpossibl | Provider model IDxai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderAuriko | Provider model IDgrok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderDaoXE | Provider model IDgrok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderCrossModel | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderAmazon Bedrock | Provider model IDxai.grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderOpper | Provider model IDxai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderxAI | Provider model IDgrok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDxai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderCloudflare AI Gateway | Provider model IDxai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderFastRouter | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderZenMux | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderOfox | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderRequesty | Provider model IDgrok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderFrogBot | Provider model IDgrok-4-3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderAIHubMix | Provider model IDgrok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDspacexai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderEden AI | Provider model IDxai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgrok-4-3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDx-ai/grok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderAbacus | Provider model IDgrok-4.3 | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context1M | Updated |
| ProviderVenice AI | Provider model IDgrok-4-3 | Regionglobal | Input / 1M$1.42 | Output / 1M$2.83 | Context1M | Updated |
Grok 4.3 Runtime Performance
Provider-specific output speed and catalog latency for Grok 4.3. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Grok 4.3 Specifications
Technical details for the model's default version.
- Version
- Grok 4.3
- Released
- May 6, 2026
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 1M
- Max output
- N/A
- Inputs
- text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Grok 4.3 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGrok 4.3 | Released | LLMBoard39.22 | ParametersN/A | Context1M | Max outputN/A | Open weightsNo | LicenseProprietary |
Models similar to Grok 4.3
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
XA
Grok Code Fast 1
xAI
XA
Grok 4.1
xAI
XA
Grok 3 Mini
xAI
XA
Grok 4.1 Thinking
xAI
XA
Grok 4 Fast
xAI
XA
Grok 3
xAI
What is Grok 4.3?
Key information about Grok 4.3 and its available data.
3 is xAI's general-purpose large language model. It supports reasoning, function calling, structured outputs, prompt caching, agentic tool calling, and instruction following.
Data as of 2026-09-08.
FAQ
Common questions about Grok 4.3.
When was Grok 4.3 released?
Grok 4.3's default version was released on May 6, 2026.
How much does Grok 4.3 cost?
Grok 4.3's official API price is $1.25 per million input tokens and $2.5 per million output tokens via xAI. The lowest tracked third-party offer starts at $1.25 input and $2.5 output via OrcaRouter.
Who created Grok 4.3?
Grok 4.3 was created by xAI.
What is the context window for Grok 4.3?
The default version has a 1M token context window.
Is Grok 4.3 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Grok 4.3?
25 provider offerings are linked to the default version.