How does Grok 4.6 rank among AI models?
Grok 4.6 is a superseded version and is not on the current board as of 28 September 2026. Its archive position, where one exists, is shown on the page; the newer version carries the current rank.
xAI·#54 of 412 in the all-versions archive
Grok 4.6 is a language model from xAI. It ranks #54 of 412 in ModelCap's all-versions archive, with an Index score of 73.6, measured on 6 public benchmarks. Its score range of 68.0–79.2 spans positions #27–#80. It is listed at $2.00 per million input tokens and $6.00 per million output tokens across 2 providers. It reads up to 500K tokens of context and is available only through an API.
Figures as of 28 Sept 2026, 02:38 UTC · download the public dataset · methodology
ModelCap Index · All-versions archive
73.6#54
Measured on 6 public benchmarks · range 68.0–79.2
Evidence as of 25 Sept 2026, 00:00 UTC
Superseded by Grok 4.7. It no longer holds a current-board position; its all-versions archive rank is #54.
Input
$2.00
per 1M tokens
Output
$6.00
per 1M tokens
Context
500K
tokens
Providers
2
5 endpoints
Providers
Uptime measured over the last 30 minutesReasoning efforts
Best supported result can inform ModelCapPublished Max, X-High, High and other configurations for this model. These are sourced scores, not head-to-head essays.
Related community configurations
Human preference ratingOther versions (4)
ModelCap Index · All-versions archive
Methodology73.6
Measured on 6 public benchmarks
All-versions archive
#54 of 412
- Score range
- 68.0–79.2
- Rank range
- #27–#80
- Evidence as of
- 25 Sept 2026, 00:00 UTC
- Benchmarks used
- ARC-AGI-2 Verified, ARC-AGI-3 Verified, Arena Coding, Arena Overall, Artificial Analysis Intelligence Index, LMArena Agent
Technical details
- Public rank basis
- Measured (measured)
- Evidence label
- Measured · 6 public benchmark observations across 6 boards
- Index support
- 88.6% · strong
- Identity binding
- Exact catalogue product · x-ai/x-ai/grok-4.6 · aggregates disclosed benchmark configurations · endpoint configuration not claimed
- Index observations used
- 6
- Qualified Index evidence point
- 73.6
Bayesian Evidence Score · measured evidence input only
BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.
- Observed capability
- 73.6
- Evidence status
- confirmed · 97% mass
- General preference
- 75.6
- Coding
- 74.4
- Agents & tools
- 44.9
- Reasoning
- 56.6
- Evidence breadth
- 100%
- Evidence coverage
- 89%
Market and catalogue signals · secondary, never the language Index rank
- Usage (OpenRouter popularity)
- 29.0
- Liquidity (providers × uptime)
- 97.6
- Open reach (HF downloads)
- —
- Surface (context / tools / modalities)
- 95.0
- Freshness
- 83.8
- · OpenRouter popularity #62
- · 2 live providers
- · Catalogue: context 500000, reasoning, tools, modalities Text/Image/File
- · First seen 2026-08-12
These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.
BES evidence selected · measured input only
- Arena Overall · High73.8
- Artificial Analysis Intelligence Index · High81.0
- Arena Coding · High74.4
- LMArena Agent · X-High44.9
- ARC-AGI-2 Verified · X-High62.5
- ARC-AGI-3 Verified · X-High43.0
The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.
Deployment readiness
Metadata index51.0
partial · 56% metadata coverage
- Artifact reproducibility
- 0.0
- Access & legal clarity
- 30.3
- Deployability
- 90.0
- Evaluation provenance
- 89.6
Not yet published: a pinned repository revision, the architecture and model type, a parameter count, safetensors weight metadata, the base model it derives from, a declared licence, a link to the licence terms, pinned evaluation dataset versions.
A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.
Market activity
Venue coverage- Market Gravity
- 52.0
- OpenRouter weekly popularity
- #62 of 334
Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.
Arena preference
- Status
- Ranked
- Source
- LMArena
- Official rank
- #72 of 409
- Rating
- 1453.4
- 95% confidence interval
- 1448–1459
- Votes
- 21K
- Source published at
- 25 Sept 2026, 00:00 UTC
- Category
- overall
- Identity match confidence
- 100%
Market Gravity breakdown
- Usage55%
- 29.0
- Liquidity25%
- 97.6
- Open reach15%
- 50.0
- Freshness5%
- 83.8
OpenRouter weekly popularity across the full model catalogue.
Independent providers versus the model's open or closed cohort, weighted by uptime.
Hugging Face 30-day downloads for open models; neutral for closed models.
Time since first public availability, on a six-month half-life.
Specification
- Model ID
x-ai/grok-4.6- Context window
- 500K tokens
- Max output
- 450K tokens
- Inputs
- Text, Image, File
- Outputs
- Text
- Tokenizer
- Grok
- Cached input tokens
- $0.50 / 1M
- First seen on OpenRouter
- 12 Aug 2026
Capabilities
- Supported: Reasoning
- Supported: Tool use
- Supported: Structured output
- Supported: Response format
- Not supported: Moderated
Benchmark leaderboards featuring Grok 4.6
All leaderboardsGrok 4.6: common questions
How much does Grok 4.6 cost per 1M tokens?
Grok 4.6 is listed at $2.00 per 1M input tokens and $6.00 per 1M output tokens as of 28 September 2026 across 2 providers. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $7.00 total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.
What is the context window of Grok 4.6?
Grok 4.6 has a published context window of 500K tokens (500,000) and a maximum output of 450K tokens. Individual providers can serve less than the published maximum; the providers table lists each endpoint's own limit.
Is Grok 4.6 open-weight?
API only. No downloadable model weights are linked in the current source data; access is through a hosted API or product.
Which API providers serve Grok 4.6?
2 providers serve Grok 4.6 through OpenRouter as of 28 September 2026: xAI and Amazon Bedrock. Each provider's price, context limit, quantization and measured uptime are in the providers table above.
Which benchmarks has Grok 4.6 been evaluated on?
Grok 4.6 has published results on 7 public boards tracked by ModelCap as of 28 September 2026: Arena coding 1507 (#59 of 404), LMArena Agent 1.2% (#23 of 44), ARC-AGI-2 67.1% (#51 of 228), ARC-AGI-3 2.1% (#23 of 57) and AA Intelligence Index 44.3 (#29 of 173) and more. Each board page ranks every tracked model on that benchmark. The ModelCap Index scores the results its methodology admits, with published uncertainty, and shows reference boards without scoring them.
When was Grok 4.6 released?
Grok 4.6 first appeared in the catalogue on 12 Aug 2026. It has since been superseded by Grok 4.7. Rank and price movements since then are recorded on the site's changes feed.