xAI Releases Grok 4.6 with Big Performance Leap at Same Price
xAI has officially released the Grok 4.6 model, with its API now available for developers. Focusing on long-horizon agentic and visual interaction capabilities, the model delivers impressive benchmark performances while maintaining the same pricing as its predecessor, drawing industry attention to xAI's rapid capability leaps.
Confirmed
- Parameters & Training: Built on a 1.5T-parameter foundation model, Grok 4.6 features major upgrades in Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL).
- Core Capability Enhancements: The model achieves significant improvements in multi-step reasoning, coding, and agentic tasks. It maintains stability in complex multi-step workflows like research analysis and codebase maintenance, while also improving complex visual interaction tasks.
- Benchmark Performance: According to official xAI data, it scored 61 in the AA Intelligence Index, tying with GPT-5.6 Sol. It also demonstrated strong performance in agentic coding and knowledge work benchmarks such as CursorBench and FrontendBench.
- API Pricing: Grok 4.6 retains the exact same pricing as the previous Grok 4.5, achieving a significant leap in frontier intelligence without passing on extra costs.
Why It Matters
- Cost-Effectiveness: By maintaining its original price point while achieving massive capability upgrades to compete with rival flagship models, it offers developers a highly cost-effective option. Industry observers have noted that xAI's pace of evolution in model capabilities has exceeded expectations.
2026-08-12 ~ 2026-08-13 · 24 related posts
- Episode 1: Grok Model Tests Show Fast Execution but Flawed Reasoning(2026-08-11, 2 posts)
- Episode 2: Grok 4.6 Spotted in Cursor Then Pulled, Unconfirmed(2026-08-11, 7 posts)
- Episode 3: Grok 4.6 Released with Multi-Tool Coding Benchmark(2026-08-11, 4 posts)
- Episode 4: Musk Confirms Grok 4.6 Release This Week Amid xAI Cursor Acquisition Talks(2026-08-12, 3 posts)
- Episode 5: xAI Releases Grok 4.6 with Big Performance Leap at Same Price(2026-08-12, 24 posts)
- Episode 6: AI Community Memes Fake Grok 4.6 Release and Benchmarks(2026-08-12, 3 posts)
- Episode 7: xAI Launches Grok 4.6 on Cursor and APIs(2026-08-12, 2 posts)
- Episode 8: Grok 4.6 Ties Claude Opus 5 Max at Top of Agent Index(2026-08-12, 2 posts)
- Episode 9: Grok 4.6 Offers 85% Cost Advantage, Next Gen to Integrate SpaceX and Cursor Data(2026-08-12, 7 posts)
Primary sources
- xAI Releases Grok 4.6 Model API — legit_api · 2026-08-12
- [source] xAI Releases Grok 4.6: Focuses on Long-Running Agents, Matches GPT-5.6 — scaling01 · 2026-08-12
- [source] xAI Launches Grok 4.6: 1.5T Parameters with Major Coding & Agent Upgrades — Daniel_Farinax · 2026-08-12
- Grok 4.6 Matches GPT-5.6 Sol on Intelligence Index, Leads in Coding Benchmarks — kimmonismus · 2026-08-12
- Grok 4.6 ties GPT-5.6 on intelligence index with top-tier agentic performance at lower cost — ArtificialAnlys · 2026-08-12
- Grok 4.6 Released, Beats GPT-5.6 on GDPVal-AA v2 Benchmark — XFreeze · 2026-08-12
- Introducing Grok 4.6: Significant Frontier Intelligence Leap at Same Price — andersonbcdefg · 2026-08-12
- Grok 4.6 Takes #1 Spot on GDPVal-AA Benchmark with 1753 Elo — elonmusk · 2026-08-12
- xAI Releases Grok 4.6: Combines Opus-Class Intelligence with Low Cost — soleio · 2026-08-12
- Elon Musk Praises Grok 4.6: Ranks #1 Across Multiple Benchmarks — elonmusk · 2026-08-12
- [source] xAI Officially Launches Grok 4.6: Major Performance Boost at the Same Price — xiaosun86 · 2026-08-12
- xAI Releases Grok 4.6, Now Available on Cursor and Grok Build — mark_k · 2026-08-12
- Grok 4.6 Tops Multiple Benchmarks While Maintaining Previous Generation Pricing — XFreeze · 2026-08-12
- Grok 4.6 Test: On Par with Sol at One-Fifth the Cost — daniel_mac8 · 2026-08-12
- Grok 4.6 Benchmarks Leaked: Major Performance Jump at Same Price — mark_k · 2026-08-13
- Grok 4.6 Hits the API at $2 per Million Input Tokens — XFreeze · 2026-08-13
- Grok 4.6 Drops and Hits LiveBench Soon, Priced Same as 4.5 — bindureddy · 2026-08-13
- Grok 4.6 Stuns in Eval: Matches Rivals at a Fraction of the Cost — daniel_mac8 · 2026-08-13
- Grok 4.6 Matches GPT-5.6 Sol in AI Index at Less Than Half the Cost — Angaisb_ · 2026-08-13
- xAI Launches Flagship Grok 4.6, Rivaling GPT-5.6 and Claude 5 at Lower Cost — 智东西 · 2026-08-13
- Grok 4.6 Scores 61 on Intelligence Index, Joining the Frontier — elonmusk · 2026-08-13
- NVIDIA Congratulates xAI on Grok 4.6: Trained on GB300 NVL72 — elonmusk · 2026-08-13
- Elon Musk Pushes Grok: Major Leap in Real-World Task Performance — elonmusk · 2026-08-13
1 near-duplicate retellings: kimmonismus