Models / Grok 4.6

xAI

Grok 4.6

Grok 4.6 is best for reasoning and knowledge, and agents and tool use; pick it when quality matters more than cost. The cheapest model of similar intelligence is Claude Haiku 5.5, at $0.20 per million tokens against $3.00.

Superseded. Grok 4.20 is xAI's newer model in this line, at $1.56 blended: see Grok 4.20 or compare the two.

Results 16 Nov 2026 · 57 results · 7 sources

Test Grok 4.6 on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$2.00 in · $6.00 out
xAI's list price, 10 Nov 2026 · compare prices$0.50 cached input
Speed
Not measured
Intelligence Index
44.3
Artificial Analysis, at high reasoning; 35.1–44.3 across 4 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
xAI, based in the US
Weights not published
Released
12 Aug 2026
500,000 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 18th of 185Claude Opus 5.5 – $3.00 blended, 73rd percentilecheapest: Mistral Nemo –
Models priced $3–10 per million tokens 10th of 38Claude Opus 5.5 – sets this group –
Models with similar intelligence sets this group – $3.00 blended, 44th percentilecheapest: Claude Haiku 5.5 –
Models from developers based in the US 14th of 92Claude Opus 5.5 – $3.00 blended, 53rd percentilecheapest: Gemma 3 4B –

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, then those priced $3–10 per million tokens. 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 Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 14th of 159 available · 10th of 35 at its price
BenchmarkGrok 4.6Rank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 21.2% 28th of 107 Kimi K3.1 Pro 67.2%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 88.4% 6th of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 50.7% 1st of 104 This model
Vending-Bench 2Vending-Bench 2: money after a year · reported $9,047 6th of 53 GPT-6 Astra $15,515
  • tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
  • Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 16th of 51 available · 9th of 23 at its price
BenchmarkGrok 4.6Rank among availableBest available
APEX-AgentsAPEX-Agents: tasks passed · reported 65.3% 6th of 37 Claude Sonnet 5.5 75.5%
GDP.pdfGDP.pdf: professional document tasks · reported 17.2% 19th of 31 GPT-5.6 Sol 30.7%
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 16th of 187 available · 10th of 39 at its price
BenchmarkGrok 4.6Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 44.3 18th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 44.1% 26th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 94.9% 3rd of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 43rd of 169 available · 19th of 34 at its price
BenchmarkGrok 4.6Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,454 48th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 63.3 28th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 55th of 139 available · 19th of 31 at its price
BenchmarkGrok 4.6Rank among availableBest available
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,449 51st of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 49.3% 25th of 68 GPT-6 Astra 75.6%
  • Vectara Hallucination Leaderboard: has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Grok 4.6 at $3.00 blended, Intelligence Index 44.3. "Similar intelligence" means within 4 points or better.

Where to run it

4 hosts on OpenRouter at the standard service tier, with a median of $3.15 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
xAInot stated$2.00$6.00$3.00
xAInot stated$2.00$6.00$3.00
Amazon Bedrocknot stated$2.20$6.60$3.30
xAInot stated$2.20$6.60$3.30

Other service tiers, not counted above: xAI priority $4.00 in, $12.00 out; xAI priority $4.00 in, $12.00 out. Flex and batch are cheaper but slower or queued; priority and fast cost more.

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

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 4.6Claude Opus 5.5
most intelligent at $3–10 per million tokens
Claude Haiku 5.5
cheapest with similar intelligence
Claude Fable 5.2
leads overall
Grok 4.7
xAI's best other model
Price per million tokens $3.00 $8.00$0.20$16.00$3.00
Tokens a second – 96240–74
Intelligence Index 44.3 57.643.4–46.4
Spring Prompt overall – 82409040
Rank among available models, by use case
Product listings – 13th20th5th9th
Decks from an analysis – 7th18th4th12th
User surveys – 4th7th1st–
Marketing planning – 9th30th1st24th
Agents and tool use 14th of 159 3rd28th–33rd
Professional work 16th of 51 3rd–––
Reasoning and knowledge 16th of 187 1st25th–23rd
Writing 43rd of 169 5th––62nd
Sticking to the facts 55th of 139 3rd––102nd
Speed under pressure – –21st8th–

Shaded figures are better than Grok 4.6's. A dash means no result.

Where it has been measured

10th most intelligent of the 38 models priced $3–10 per million tokens. Among models available today, in the top quarter (by percentile) for reasoning and knowledge, and agents and tool use.

Artificial Analysis lists it but has not published a speed for it yet

Every result

57 published results from 7 sources, each in the source's own units, with the configuration that produced it.

APEX-Agents · reported by APEX-Agents · 1 result
MeasureValueRankConfigurationDated
APEX-Agents: tasks passed% of tasks, higher is better 65.3% 6 of 39 Grok 4.6 1 Oct 2026

Not shown: Not your firm's documents or tools; graded by rubric, not by a client.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 12 results
MeasureValueRankConfigurationDated
Arena Agent: confirmed task successIPS effect estimate, higher is better -0.01
-0.03–0.02
31 of 50 Grok 4.6 (xhigh reasoning, arena agent) 2 Oct 2026
Arena Agent: praise over complaintIPS effect estimate, higher is better 0.00
-0.03–0.04
28 of 50 Grok 4.6 (xhigh reasoning, arena agent) 2 Oct 2026
Arena Agent: steerabilityIPS effect estimate, higher is better 0.01
-0.01–0.02
31 of 50 Grok 4.6 (xhigh reasoning, arena agent) 2 Oct 2026
Arena Agent: tool groundingIPS effect estimate, higher is better 0.00
0.00–0.00
21 of 50 Grok 4.6 (xhigh reasoning, arena agent) 2 Oct 2026
Arena Document: overallArena rating, higher is better 1,461
1,449–1,473
24 of 44 Grok 4.6 (high reasoning) 13 Sep 2026
Arena Text factuality: overallArena rating, higher is better 1,449
1,444–1,453
71 of 182 Grok 4.6 (high reasoning) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,447
1,437–1,456
83 of 406 Grok 4.6 (high reasoning) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,445
1,436–1,454
52 of 411 Grok 4.6 (high reasoning) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,491
1,479–1,503
59 of 363 Grok 4.6 (high reasoning) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,451
1,443–1,458
63 of 413 Grok 4.6 (high reasoning) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,454
1,449–1,459
73 of 413 Grok 4.6 (high reasoning) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,445
1,437–1,454
62 of 412 Grok 4.6 (high reasoning) 2 Oct 2026

Not shown: Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.

Artificial Analysis · reported by Artificial Analysis · 36 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 76.8 12 of 187 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 66.3 64 of 187 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 74.4 30 of 187 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 75.9 23 of 187 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 44.3 39 of 413 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 35.1 85 of 413 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 42.8 45 of 413 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 44.2 40 of 413 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 94.9% 4 of 360 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 87.9% 95 of 360 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 93.5% 14 of 360 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 93.5% 14 of 360 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 42.9% 63 of 419 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 27.6% 169 of 419 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 42.1% 75 of 419 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 44.1% 52 of 419 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 80.3% 70 of 407 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 80.7% 64 of 407 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 81.0% 58 of 407 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 81.0% 58 of 407 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 56.5% 35 of 205 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 49.4% 119 of 205 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 55.9% 41 of 205 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 53.0% 81 of 205 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 88.4% 10 of 188 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 75.3% 63 of 188 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 84.3% 30 of 188 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 88.0% 12 of 188 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 21.2% 66 of 204 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 3.0% 116 of 204 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 13.1% 84 of 204 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 17.2% 73 of 204 Grok 4.6 (xhigh reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 50.7% 2 of 175 Grok 4.6 (high reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 38.1% 38 of 175 Grok 4.6 (low reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 44.3% 18 of 175 Grok 4.6 (medium reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 43.3% 20 of 175 Grok 4.6 (xhigh reasoning) 20 Oct 2026

Not shown: Not your codebase or tools.

GDP.pdf · reported by GDP.pdf · 2 results
MeasureValueRankConfigurationDated
GDP.pdf: professional document tasks% of rubric, higher is better 16.0% 31 of 49 Grok 4.6 (high reasoning) 1 Oct 2026
GDP.pdf: professional document tasks% of rubric, higher is better 17.2% 27 of 49 Grok 4.6 (xhigh reasoning) 1 Oct 2026

Not shown: Not your documents; 100 tasks, so small differences are noise.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 2 results
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 49.3% 26 of 85 Grok 4.6 (high reasoning) 27 Aug 2026
SimpleQA Verified: correct answers% of questions, higher is better 48.9% 27 of 85 Grok 4.6 (xhigh reasoning) 27 Aug 2026

Not shown: Not answers grounded in your documents; tests what the model remembers.

UGI Leaderboard · reported by UGI Leaderboard · 3 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 25.0% 236 of 379 Grok 4.6 3 Sep 2026
UGI: style adherencescore from 0 to 1, higher is better 0.36 149 of 379 Grok 4.6 3 Sep 2026
UGI: writing scorescore out of 100, higher is better 63.3 82 of 379 Grok 4.6 3 Sep 2026

Not shown: Not other format limits such as character counts or bullet counts.

Vending-Bench 2 · reported by Vending-Bench 2 · 1 result
MeasureValueRankConfigurationDated
Vending-Bench 2: money after a yearUS dollars, higher is better $9,047 6 of 63 Grok 4.6 1 Oct 2026

Not shown: Not a real business; one simulated market with set rules.

Will Grok 4.6 work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.