Models / Grok 4
Grok 4
Early evidence, 8 results from 2 sources, on 1 use case. Grok 4 is strongest at reasoning and knowledge; it is not on sale through OpenRouter, so it is here for reference.
Superseded. Grok 4.20 is xAI's newer model in this line, at $1.56 blended: see Grok 4.20.
Results 16 Nov 2026 · 8 results · 2 sources
- Price per million tokens
- Not available through OpenRouter
- Speed
- Not measured
- Intelligence Index
- 22.5
Artificial Analysis - Spring Prompt overall
- Not ranked yet
needs our benchmarks and two groups of results - Developer
- xAI, based in the US
Weights not published - Released
- 9 Jul 2025
Where it stands among the models you could choose
Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.
| Among | Intelligence Index | Spring Prompt overall | Price, blended | Speed |
|---|---|---|---|---|
| Models available today | 88th of 186Claude Opus 5.5 | – | – | – |
| Models from developers based in the US | 51st of 93Claude Opus 5.5 | – | – | – |
The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.
How good is it, and for what?
Each use case ranks the available models its benchmarks measured. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.
Product listingsTurning a sparse product feed and photos into listings that can go live
- CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is
- DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled
- SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending
- ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
- tau2-bench: has not measured this model.
- Berkeley Function Calling Leaderboard (BFCL) V4: has not measured this model.
- OpenHands Index: has not measured this model.
- Microsoft STATE-Bench: has not measured this model.
- Artificial Analysis: has not measured this model.
- Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects
- APEX-Agents: has not measured this model.
- GDP.pdf: has not measured this model.
- Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge
| Benchmark | Grok 4 | Rank among available | Best available |
|---|---|---|---|
| Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported | 22.5 | 88th of 186 | Claude Opus 5.5 57.6 |
| Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported | 26.7% | 89th of 187 | Claude Opus 5.5 61.4% |
| Artificial AnalysisArtificial Analysis: GPQA Diamond · reported | 87.7% | 55th of 169 | GPT-6 Astra 96.3% |
WritingWhat people prefer in blind comparisons, and judged writing quality
- Arena (formerly LMArena): has not measured this model.
- UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers
- Vectara Hallucination Leaderboard: has not measured this model.
- Arena (formerly LMArena): has not measured this model.
- SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess)
- BulletBench: has not measured this model.
Nearest alternatives
- XAI's newer model in this lineGrok 4.20$1.56 blended · Intelligence Index 25.7
- Open weights at similar intelligenceDeepSeek-V4-Flash (0423)$0.15 blended · Intelligence Index 24.4Compare with Grok 4 →
Against Grok 4, Intelligence Index 22.5. "Similar intelligence" means within 4 points or better.
Against the alternatives
The models you would most likely weigh it against: the leaders of the groups above.
Scroll sideways to see every alternative.
| Grok 4 | Claude Opus 5.5 most intelligent available today | Claude Fable 5.2 leads overall | Grok 4.7 xAI's best other model | GPT-6.1 Sol near the top overall | |
|---|---|---|---|---|---|
| Price per million tokens | – | $8.00 | $16.00 | $3.00 | $4.00 |
| Tokens a second | – | 96 | – | 74 | 53 |
| Intelligence Index | 22.5 | 57.6 | – | 46.4 | 51.8 |
| Spring Prompt overall | – | 82 | 90 | 40 | 80 |
| Rank among available models, by use case | |||||
| Product listings | – | 13th | 5th | 9th | 2nd |
| Decks from an analysis | – | 7th | 4th | 12th | 5th |
| User surveys | – | 4th | 1st | – | 3rd |
| Marketing planning | – | 9th | 1st | 24th | – |
| Agents and tool use | – | 3rd | – | 33rd | 6th |
| Professional work | – | 3rd | – | – | 9th |
| Reasoning and knowledge | 84th of 188 | 1st | – | 23rd | 6th |
| Writing | – | 5th | – | 62nd | 16th |
| Sticking to the facts | – | 3rd | – | 102nd | 18th |
| Speed under pressure | – | – | 8th | – | 3rd |
| Side by side → | Side by side → | Side by side → | |||
Shaded figures are better than Grok 4's. A dash means no result.
Where it has been measured
Compared with the models available today (it is not on sale itself), mid-field on reasoning and knowledge, the one use case measured.
Artificial Analysis lists it but has not published a speed for it yet
- Arena (formerly LMArena)1 result
- Artificial Analysis7 results
- BulletBenchNot measured
- CatalogBenchNot measured
- DeckBenchNot measured
- ROASBenchNot measured
- SurveyBenchNot measured
- APEX-AgentsNot measured
- Berkeley Function Calling Leaderboard (BFCL) V4Not measured
- GDP.pdfNot measured
- Microsoft STATE-BenchNot measured
- OpenHands IndexNot measured
- Remote Labor IndexNot measured
- SimpleQA Verified (Epoch AI)Not measured
- tau2-benchNot measured
- UGI LeaderboardNot measured
- Vectara Hallucination LeaderboardNot measured
- Vending-Bench 2Not measured
Every result
8 published results from 2 sources, each in the source's own units, with the configuration that produced it.
Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 1 result
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| Arena Search: overallArena rating, higher is better | 1,121 1,115–1,128 |
32 of 34 | Grok 4 (arena search) | 24 Aug 2026 |
Not shown: Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Artificial Analysis · reported by Artificial Analysis · 7 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better | 22.5 | 184 of 413 | Grok 4 | 20 Oct 2026 |
| Artificial Analysis: GPQA Diamond% of questions, higher is better | 87.7% | 97 of 360 | Grok 4 | 20 Oct 2026 |
| Artificial Analysis: Humanity's Last Exam% of questions, higher is better | 26.7% | 173 of 419 | Grok 4 | 20 Oct 2026 |
| Artificial Analysis: IFBench% of instructions, higher is better | 53.7% | 126 of 288 | Grok 4 | 20 Oct 2026 |
| Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better | 68.0% | 215 of 407 | Grok 4 | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better | 37.9% | 60 of 282 | Grok 4 | 20 Oct 2026 |
| Artificial Analysis: τ²-bench telecom% of tasks, higher is better | 74.9% | 114 of 286 | Grok 4 | 20 Oct 2026 |
Not shown: Not business work, and a blend: read the parts for any one task.
Test your own prompt.
Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.