Released August 12, 2026, succeeding Grok 4.5. Long-context billing is a cliff, not a ramp: at 200K tokens and above the whole request is charged at $4/$12. DeepSWE 65.9%, CursorBench 3.2 70.8%, FrontierCode 1.1 Extended 61.3%, Terminal-Bench 3.0 26.5%, APEX-Agents 57.5%.
Roughly half the turns of competing models on agent tasks — real cost per completed task lands well below the sticker price
88.4% on Terminal-Bench 2.1 and 87.0% on VulcanBench v3
Artificial Analysis Intelligence Index of 61, 4th overall and within reach of Claude Opus 5
Weaknesses
No published SWE-bench Verified or SWE-bench Pro figure, so it cannot be compared directly on the standard coding leaderboard
GPT-5.6 Sol Max beats it on DeepSWE v1.1 and Terminal-Bench 3.0
Prompts of 200K tokens or more are billed at $4/$12 across the entire request, not just the overage
Real-world use cases
What people actually use Grok 4.6 for.
Long-horizon research agents that work a topic across many sequential steps
Working through an unfamiliar codebase over an extended agent session
Interactive and visual work where turn count drives the real bill
How Grok 4.6 compares
The nearest models people weigh against it, and what actually separates them.
vs Grok 4.7 — Against Grok 4.7 (xAI), Grok 4.6 lands within a few percent on price. Which one wins depends on whether context depth or latency is your constraint.
vs Claude 3.5 Sonnet — Against Claude 3.5 Sonnet (Anthropic), Grok 4.6 runs about 78% cheaper per token, takes 2.5x the context and answers faster. Take Grok 4.6 unless you specifically need what Claude 3.5 Sonnet does better.
vs Claude 3.7 Sonnet (thinking) — Against Claude 3.7 Sonnet (thinking) (Anthropic), Grok 4.6 runs about 56% cheaper per token, takes 2.5x the context and answers faster. Take Grok 4.6 unless you specifically need what Claude 3.7 Sonnet (thinking) does better.
Price History
Grok 4.6 pricing over time
→0% since Sep 1
43 data points · tracked daily since Sep 1, 2026
Ready to try it?
Start using Grok 4.6
Long-running agents and multi-step codebase work. Start free — no card required.
Grok 4.7, released September 21, 2026, targets coding and professional knowledge work over multi-hour tasks, with better self-verification and long-context handling than Grok 4.6 at the same $2/$6 price.
Verdict
Grok 4.6's price with modest coding gains and heavier token use.
Quality score
84%
Pricing
$2.00/1M in
$6.00/1M out
Speed
Fast
4/5 speed
Context
500k tokens
Released September 21, 2026. Gateway id spacexai/grok-4.7. $2/$6 per 1M up to 200K input tokens, $4/$12 above. 500,000 context, 500,000 max output, knowledge cutoff May 2026. Vendor launch figures: CursorBench 4.0 46.3 (xhigh); Terminal-Bench 4.0 37.6; DeepSWE v1.1 71.0 (high). Third-party: Senior SWE-Bench pass@3 40.0 (Snorkel AI, Sep 22, 2026); SWE-Together pass@1 64.7. Verified October 10, 2026.
AgenticCodingReasoningNew
Best for
Multi-hour coding and document tasks at mid-tier pricing
Claude 3.5 Sonnet is Anthropic's mid-cycle flagship model, balancing strong reasoning, coding, and instruction-following with a 200K context window. It sits between Haiku and Opus in Anthropic's lineup, offering near-flagship quality at a lower cost than top-tier models.
Verdict
One of the best models for coding and complex instruction-following, but its premium pricing demands premium use cases.
Quality score
81%
Pricing
$6.00/1M in
$30.00/1M out
Speed
Balanced
3/5 speed
Context
200k tokens
Pricing at $6 input / $30 output per million tokens is significantly higher than GPT-4o ($2.50/$10). Best accessed via Anthropic API or Amazon Bedrock. Claude 3.5 Sonnet (October 2024 version) supersedes the June 2024 release with improved performance.
Claude 3.7 Sonnet with extended thinking enabled — Anthropic's hybrid reasoning model that explicitly deliberates before responding, surfacing its chain-of-thought for complex multi-step problems. It sits between standard Sonnet and full reasoning-only models, balancing depth with practical usability.
Verdict
The most transparent reasoning model on the market — ideal when you need to see and trust the thought process, not just the answer.
Quality score
73%
Pricing
$3.00/1M in
$15.00/1M out
Speed
Deliberate
2/5 speed
Context
200k tokens
Thinking tokens (the internal reasoning trace) count toward output token billing, which can significantly increase costs on complex queries. The thinking budget can often be configured via the API. Best used selectively for tasks that genuinely benefit from deliberation rather than as a default model.
ReasoningExtended ThinkingCodingAgenticAnthropic
Best for
Tackling complex coding challenges, mathematical proofs, and multi-step logical problems where visible reasoning and higher accuracy matter more than speed.
Grok 4.6 costs $2 per million input tokens and $6 per million output tokens on the API, with cached input at $0.5 per million. A month of 10M input and 2M output tokens runs about $32.00 at list price, before any batch or caching discounts.
What is the context window of Grok 4.6?
Grok 4.6 has a 500k tokens context window, with up to 500k tokens of output per response. That is the total of prompt plus response the model can hold in one request.
What is the knowledge cutoff of Grok 4.6?
Grok 4.6's training data runs through February 1, 2026, and the model was released on August 12, 2026. For anything after that date it needs web search or documents in the prompt.
What is Grok 4.6 best for?
Grok 4.6 is best for long-running agents and multi-step codebase work. It is a strong fit when that workflow matters more than the tradeoffs around balanced pricing and fast speed.
When should I avoid Grok 4.6?
You need a published SWE-bench score to justify the pick, or your prompts routinely cross 200K tokens where the price doubles.
What is a cheaper alternative to Grok 4.6?
Claude Sonnet 5.5 (Anthropic) at $2.00/1M/1M input against Grok 4.6's $2.00/1M/1M. Near-Opus 5.5 quality on scoped work at half the price. Compare it first if Grok 4.6's pricing is the thing stopping you.
What is a faster alternative to Grok 4.6?
Grok 4.7 — fast against Grok 4.6's fast, with 500k tokens of context. Worth the swap when response time is what your users notice rather than the last few points of reasoning depth.
Newsletter
Get notified when Grok 4.6 pricing changes
We track pricing daily. When this model drops or spikes, you'll know first.
No spam. Useful updates only. Affiliate disclosures always clearly labeled.