# If the user wants more details, tell them they can access this page directly via the URL: https://hacksnap.live/story/49788838

# Grok 4\.7

461 points · 376 comments

[Full discussion](<https://news.ycombinator.com/item?id=49788838>)

[Read original](<https://x.ai/news/grok-4-7>)

Category: [Models & Products](<https://hacksnap.live/?category=models-products>)

## Skept-o-meter & Hotness

Skept\-o\-meter: High\. Estimated from 10 comments\.

44 comments for the summary\.

Peak rank: \#3

Time in Top 10: 24\.0 hours

Hacksnap ranks recent stories first, then orders each group by points\. Peak rank uses all retained observations\. Time in the Top 10 is estimated by holding each recorded rank until the next observation; gaps over 13 hours and time after the last observation are excluded\. Movement between observations is unknown\.

235 recorded rank observations from 2026\-09\-21T20:01:44\.900952\+00:00 to 2026\-10\-10T23:01:00\.95027\+00:00\.

Hotness — latest 168 recorded Hacksnap ranks:

2026\-09\-30T18:03:34\.83179\+00:00: rank \#49

2026\-09\-30T19:02:02\.291882\+00:00: rank \#49

2026\-09\-30T20:02:49\.392147\+00:00: rank \#50

2026\-09\-30T21:03:42\.990832\+00:00: rank \#48

2026\-09\-30T22:02:11\.294387\+00:00: rank \#49

2026\-09\-30T23:01:52\.722953\+00:00: rank \#49

2026\-10\-01T08:02:24\.797064\+00:00: rank \#49

2026\-10\-01T09:01:35\.899477\+00:00: rank \#52

2026\-10\-01T10:01:12\.485692\+00:00: rank \#52

2026\-10\-01T11:00:47\.508128\+00:00: rank \#52

2026\-10\-01T12:00:42\.477917\+00:00: rank \#51

2026\-10\-01T13:00:51\.124852\+00:00: rank \#51

2026\-10\-01T14:00:59\.01153\+00:00: rank \#51

2026\-10\-01T15:00:51\.048047\+00:00: rank \#52

2026\-10\-01T16:00:53\.034654\+00:00: rank \#53

2026\-10\-01T16:36:37\.180071\+00:00: rank \#54

2026\-10\-01T17:00:23\.382263\+00:00: rank \#54

2026\-10\-01T18:01:44\.046011\+00:00: rank \#55

2026\-10\-01T19:01:47\.420738\+00:00: rank \#56

2026\-10\-01T20:01:39\.585083\+00:00: rank \#55

2026\-10\-01T21:01:45\.872085\+00:00: rank \#56

2026\-10\-01T22:01:37\.214899\+00:00: rank \#55

2026\-10\-01T23:04:11\.395288\+00:00: rank \#58

2026\-10\-02T08:02:26\.819307\+00:00: rank \#59

2026\-10\-02T09:02:51\.51649\+00:00: rank \#60

2026\-10\-02T10:02:12\.346099\+00:00: rank \#60

2026\-10\-02T11:02:22\.794376\+00:00: rank \#60

2026\-10\-02T12:01:14\.681578\+00:00: rank \#60

2026\-10\-02T13:02:29\.511591\+00:00: rank \#60

2026\-10\-02T14:01:40\.936738\+00:00: rank \#59

2026\-10\-02T15:02:30\.225476\+00:00: rank \#59

2026\-10\-02T16:02:47\.060695\+00:00: rank \#58

2026\-10\-02T17:02:31\.351493\+00:00: rank \#60

2026\-10\-02T18:01:22\.271899\+00:00: rank \#61

2026\-10\-02T19:01:23\.681417\+00:00: rank \#61

2026\-10\-02T20:01:33\.801055\+00:00: rank \#63

2026\-10\-02T21:02:55\.453316\+00:00: rank \#66

2026\-10\-02T22:01:50\.233139\+00:00: rank \#67

2026\-10\-02T23:01:20\.268516\+00:00: rank \#65

2026\-10\-03T08:01:30\.186188\+00:00: rank \#66

2026\-10\-03T09:01:57\.91467\+00:00: rank \#64

2026\-10\-03T10:00:55\.732001\+00:00: rank \#63

2026\-10\-03T11:01:22\.538596\+00:00: rank \#63

2026\-10\-03T12:01:08\.789384\+00:00: rank \#64

2026\-10\-03T13:00:35\.273559\+00:00: rank \#64

2026\-10\-03T14:00:48\.425606\+00:00: rank \#64

2026\-10\-03T15:01:04\.472963\+00:00: rank \#64

2026\-10\-03T16:01:56\.56175\+00:00: rank \#65

2026\-10\-03T17:01:36\.651036\+00:00: rank \#62

2026\-10\-03T18:01:58\.24207\+00:00: rank \#62

2026\-10\-03T19:01:19\.196589\+00:00: rank \#61

2026\-10\-03T20:00:57\.313759\+00:00: rank \#59

2026\-10\-03T21:01:04\.44439\+00:00: rank \#57

2026\-10\-03T21:20:01\.585706\+00:00: rank \#56

2026\-10\-03T22:00:15\.503126\+00:00: rank \#56

2026\-10\-03T23:00:55\.584832\+00:00: rank \#56

2026\-10\-04T08:01:00\.327459\+00:00: rank \#55

2026\-10\-04T09:00:59\.321023\+00:00: rank \#56

2026\-10\-04T10:00:42\.975336\+00:00: rank \#56

2026\-10\-04T11:01:03\.76577\+00:00: rank \#56

2026\-10\-04T12:01:45\.307985\+00:00: rank \#56

2026\-10\-04T13:00:49\.176267\+00:00: rank \#56

2026\-10\-04T14:00:56\.952543\+00:00: rank \#56

2026\-10\-04T15:00:54\.739637\+00:00: rank \#56

2026\-10\-04T16:00:51\.403054\+00:00: rank \#57

2026\-10\-04T17:00:43\.166713\+00:00: rank \#57

2026\-10\-04T18:00:46\.315535\+00:00: rank \#56

2026\-10\-04T19:00:54\.768841\+00:00: rank \#56

2026\-10\-04T20:00:51\.070442\+00:00: rank \#57

2026\-10\-04T21:00:28\.550532\+00:00: rank \#56

2026\-10\-04T22:01:36\.217651\+00:00: rank \#58

2026\-10\-04T23:01:09\.400767\+00:00: rank \#58

2026\-10\-05T08:01:51\.387003\+00:00: rank \#57

2026\-10\-05T09:01:40\.528475\+00:00: rank \#56

2026\-10\-05T10:00:56\.290772\+00:00: rank \#56

2026\-10\-05T11:01:05\.919518\+00:00: rank \#56

2026\-10\-05T12:00:56\.590881\+00:00: rank \#55

2026\-10\-05T13:01:05\.197156\+00:00: rank \#57

2026\-10\-05T14:01:04\.085718\+00:00: rank \#57

2026\-10\-05T15:00:50\.814571\+00:00: rank \#57

2026\-10\-05T16:02:09\.54139\+00:00: rank \#56

2026\-10\-05T17:00:35\.336076\+00:00: rank \#56

2026\-10\-05T18:00:20\.445491\+00:00: rank \#56

2026\-10\-05T19:02:43\.604975\+00:00: rank \#57

2026\-10\-05T20:01:09\.777459\+00:00: rank \#57

2026\-10\-05T21:01:47\.650671\+00:00: rank \#58

2026\-10\-05T22:02:59\.93099\+00:00: rank \#58

2026\-10\-06T08:02:11\.210914\+00:00: rank \#58

2026\-10\-06T08:02:14\.354697\+00:00: rank \#58

2026\-10\-06T09:01:21\.688765\+00:00: rank \#60

2026\-10\-06T10:00:58\.047773\+00:00: rank \#60

2026\-10\-06T11:00:41\.689929\+00:00: rank \#61

2026\-10\-06T12:00:41\.7705\+00:00: rank \#62

2026\-10\-06T13:01:24\.509519\+00:00: rank \#62

2026\-10\-06T14:00:26\.950817\+00:00: rank \#62

2026\-10\-06T15:00:49\.066414\+00:00: rank \#63

2026\-10\-06T16:00:58\.166747\+00:00: rank \#65

2026\-10\-06T17:01:17\.009511\+00:00: rank \#65

2026\-10\-06T18:01:43\.791429\+00:00: rank \#66

2026\-10\-06T19:00:46\.048609\+00:00: rank \#66

2026\-10\-06T20:01:14\.687294\+00:00: rank \#66

2026\-10\-06T21:02:13\.414382\+00:00: rank \#66

2026\-10\-06T22:00:40\.926271\+00:00: rank \#65

2026\-10\-06T23:03:26\.867445\+00:00: rank \#67

2026\-10\-07T08:01:41\.339804\+00:00: rank \#68

2026\-10\-07T09:02:38\.026986\+00:00: rank \#69

2026\-10\-07T10:01:35\.069531\+00:00: rank \#69

2026\-10\-07T11:02:07\.770084\+00:00: rank \#69

2026\-10\-07T12:01:14\.996697\+00:00: rank \#68

2026\-10\-07T13:00:47\.45912\+00:00: rank \#67

2026\-10\-07T14:01:18\.307739\+00:00: rank \#67

2026\-10\-07T15:01:57\.541362\+00:00: rank \#69

2026\-10\-07T16:01:28\.474022\+00:00: rank \#68

2026\-10\-07T17:01:15\.036908\+00:00: rank \#69

2026\-10\-07T18:00:53\.300079\+00:00: rank \#68

2026\-10\-07T19:01:49\.456657\+00:00: rank \#69

2026\-10\-07T20:01:59\.032487\+00:00: rank \#69

2026\-10\-07T21:05:36\.655392\+00:00: rank \#71

2026\-10\-07T22:02:27\.105241\+00:00: rank \#72

2026\-10\-07T23:01:17\.452894\+00:00: rank \#71

2026\-10\-08T08:03:55\.083808\+00:00: rank \#71

2026\-10\-08T09:03:58\.68898\+00:00: rank \#69

2026\-10\-08T10:04:05\.364235\+00:00: rank \#70

2026\-10\-08T11:02:04\.410307\+00:00: rank \#69

2026\-10\-08T12:02:36\.39618\+00:00: rank \#69

2026\-10\-08T13:03:09\.206401\+00:00: rank \#69

2026\-10\-08T14:03:30\.113384\+00:00: rank \#70

2026\-10\-08T15:02:54\.444758\+00:00: rank \#68

2026\-10\-08T16:03:13\.067807\+00:00: rank \#67

2026\-10\-08T17:03:47\.92465\+00:00: rank \#66

2026\-10\-08T18:01:50\.360431\+00:00: rank \#66

2026\-10\-08T19:02:55\.719274\+00:00: rank \#67

2026\-10\-08T20:01:36\.359022\+00:00: rank \#67

2026\-10\-08T21:03:34\.000719\+00:00: rank \#67

2026\-10\-08T22:01:13\.459732\+00:00: rank \#66

2026\-10\-08T23:01:25\.318392\+00:00: rank \#65

2026\-10\-09T08:01:53\.314021\+00:00: rank \#66

2026\-10\-09T09:02:59\.93426\+00:00: rank \#67

2026\-10\-09T10:02:22\.652527\+00:00: rank \#66

2026\-10\-09T11:02:04\.807282\+00:00: rank \#66

2026\-10\-09T12:02:16\.769295\+00:00: rank \#66

2026\-10\-09T13:02:30\.35034\+00:00: rank \#68

2026\-10\-09T14:01:34\.215134\+00:00: rank \#67

2026\-10\-09T15:01:27\.564392\+00:00: rank \#66

2026\-10\-09T16:02:06\.561056\+00:00: rank \#67

2026\-10\-09T17:03:06\.673648\+00:00: rank \#69

2026\-10\-09T18:01:02\.851052\+00:00: rank \#69

2026\-10\-09T19:01:11\.250291\+00:00: rank \#69

2026\-10\-09T20:01:32\.91907\+00:00: rank \#69

2026\-10\-09T21:02:08\.376144\+00:00: rank \#71

2026\-10\-09T22:01:19\.253075\+00:00: rank \#73

2026\-10\-09T23:01:31\.990912\+00:00: rank \#73

2026\-10\-10T08:01:00\.593376\+00:00: rank \#72

2026\-10\-10T09:01:45\.527975\+00:00: rank \#74

2026\-10\-10T10:00:53\.799353\+00:00: rank \#74

2026\-10\-10T11:00:35\.65851\+00:00: rank \#73

2026\-10\-10T12:01:11\.169647\+00:00: rank \#73

2026\-10\-10T13:00:54\.008377\+00:00: rank \#71

2026\-10\-10T14:01:43\.208766\+00:00: rank \#73

2026\-10\-10T15:01:24\.825188\+00:00: rank \#73

2026\-10\-10T16:01:10\.213328\+00:00: rank \#72

2026\-10\-10T17:00:48\.84189\+00:00: rank \#71

2026\-10\-10T18:00:57\.55132\+00:00: rank \#71

2026\-10\-10T19:00:52\.379463\+00:00: rank \#70

2026\-10\-10T20:00:21\.993912\+00:00: rank \#70

2026\-10\-10T21:01:10\.657738\+00:00: rank \#70

2026\-10\-10T22:00:50\.433569\+00:00: rank \#69

2026\-10\-10T23:01:00\.95027\+00:00: rank \#70

The announcement positions Grok 4\.7 as a price\-performance and safety advance, but the sampled discussion is largely skeptical: commenters question benchmark comparability and omissions, report token\-efficiency and hands\-on reliability problems, note missing cached\-input pricing, prefer cheaper or rival models, and raise political and UX objections\.

## The brief

xAI announces Grok 4\.7 as its most capable model for coding and knowledge work, claiming longer work on difficult tasks, better self\-checking, improved long\-context handling, and its best\-calibrated safeguards\. It is served at the same price and speed as Grok 4\.6 and is positioned as highly competitive on price\-performance, with benchmark comparisons against Fable 5\.1, Opus 5, GPT\-5\.6 Sol, Sonnet 5, and Grok 4\.6\.

- Grok 4\.7 is described as xAI's most capable model for coding and knowledge work, working longer on difficult tasks, checking its own work more carefully, and coming with the best\-calibrated safeguards to date\.
- It uses a new, larger base model than Grok 4\.6, trained with a longer reinforcement learning run on a harder task mix weighted toward multi\-hour problems, with better self\-verification, long\-context management, and native understanding of the Grok Bot harness\.
- xAI claims frontier price\-performance on CursorBench 4\.0 and reports scores including CursorBench 46\.3%, DeepSWE 71\.0% at high effort, EEBench 64\.0%, AA Briefcase 1,657, Terminal\-Bench 38\.0%, Harvey Legal Agent Benchmark 19\.6%, and HealthBench Professional 56\.7%\.
- Safety and cybersecurity claims include an entirely new safeguard stack, strongest tested refusal and jailbreak resistance, 62\.4% on LatchBio's biosafety benchmark, and HackerBench v0\.3 allowing only 3\.3% of risky dual\-use prompts through while rarely blocking legitimate security work\.
- Pricing and availability: starting at $2 per million input tokens and $6 per million output tokens, with a fast variant at twice the output speed and twice the price, available in Cursor, Grok Build, the Grok API, third\-party coding harnesses, model routers, and cloud platforms\.
- The announcement includes charts comparing Fable 5\.1, Opus 5, Grok 4\.7, GPT\-5\.6 Sol, and Sonnet 5 on average cost per task, plus GDPval, AA Briefcase, and EEBench comparisons against Grok 4\.6, Fable 5\.1, and GPT\-6 Astra\.

## Discussion themes

### Hands\-on reliability concerns

Saline9515 reports that Grok 4\.7 in Omp loops in thinking mode, ignores AGENTS\.md instructions, and corrupts plan files, saying Grok 4\.6 was not so bad\. xmorse dismisses OMP itself as a joke, challenging the validity of that test setup\.

Sources: [Comment 49789800](<https://news.ycombinator.com/item?id=49789800>) · [Comment 49790162](<https://news.ycombinator.com/item?id=49790162>)

### Benchmark fairness and omissions

AM1010101 asks why Grok 4\.7 x\-high is compared with Grok 4\.6 high, and ssutch3 notes xhigh is new to Grok\. kristofferR calls the top graph deceptive for omitting Astra, while Jcampuzano2 explains that OpenAI does not allow Astra in Cursor and CursorBench uses Cursor as the harness\.

Sources: [Comment 49789853](<https://news.ycombinator.com/item?id=49789853>) · [Comment 49790165](<https://news.ycombinator.com/item?id=49790165>) · [Comment 49789298](<https://news.ycombinator.com/item?id=49789298>) · [Comment 49789424](<https://news.ycombinator.com/item?id=49789424>) · [Comment 49789418](<https://news.ycombinator.com/item?id=49789418>)

### Cost, alternatives, and pricing details

meerita says Grok is really expensive and gets amazing results from DeepSeek 4\.1 Flash for a fraction of the price\. saejox says Grok 4\.7 is not close to Astra, which uses fewer tokens, and that xAI missed its chance\. jasonjmcghee prefers Sol and Astra to Opus and Fable on cost payoff, and simonw notes no cached input token pricing was visible\.

Sources: [Comment 49789908](<https://news.ycombinator.com/item?id=49789908>) · [Comment 49790166](<https://news.ycombinator.com/item?id=49790166>) · [Comment 49789415](<https://news.ycombinator.com/item?id=49789415>) · [Comment 49790059](<https://news.ycombinator.com/item?id=49790059>)

### Token efficiency and reasoning behavior

notduckrabbit cites artificialanalysis\.ai as suggesting a significant token\-efficiency regression compared with Grok 4\.6\. simonw reports odd reasoning\-token counts across effort levels, with little difference between levels and xhigh sometimes using fewer reasoning tokens than high\.

Sources: [Comment 49790640](<https://news.ycombinator.com/item?id=49790640>) · [Comment 49790209](<https://news.ycombinator.com/item?id=49790209>)

### Language, personality, and user experience

simianwords says Grok is the worst among major models for personal chat, with a bland personality that does not work hard or try to help\. artemonster counters that Grok does good research and produces less bullshit when prompted to be critical\. moojacob likes the new Groks' plain English and dislikes Claudish, Capricorn2481 questions whether LLM personality is desirable, and smashers1114 suggests ASD\-STE100 as a workaround for Claudish\.

Sources: [Comment 49789163](<https://news.ycombinator.com/item?id=49789163>) · [Comment 49789380](<https://news.ycombinator.com/item?id=49789380>) · [Comment 49789234](<https://news.ycombinator.com/item?id=49789234>) · [Comment 49789448](<https://news.ycombinator.com/item?id=49789448>) · [Comment 49790348](<https://news.ycombinator.com/item?id=49790348>)

### Political and ethical baggage

dom96 says negative political baggage led them to exclude Grok from their LLM benchmarks\. thih9 refuses to use Grok, citing the usual reasons and saying xAI does not seem to care about user experience or long\-term support\. mlindner complains about Musk\-related political rants in product threads, and big\_toast hopes HN continues to surface disagreement and substantial views\.

Sources: [Comment 49790250](<https://news.ycombinator.com/item?id=49790250>) · [Comment 49790572](<https://news.ycombinator.com/item?id=49790572>) · [Comment 49790367](<https://news.ycombinator.com/item?id=49790367>) · [Comment 49790837](<https://news.ycombinator.com/item?id=49790837>)

## Sources & coverage

AI-generated summary · 2026\-09\-21T20:01:06\.589722\+00:00

Based on 44 of 44 usable stored comments, selected by depth and branch activity. This is a sample of the discussion. Article text may also be shortened.

Generated using deepseek\-ai/DeepSeek\-V4\.1\-Flash. Check the linked sources for full context.