{"slug": "grok-fast", "name": "Grok Fast", "category": "ai-models", "category_name": "AI models", "description": "", "recommenders": 4, "commenters": 8, "warners": 2, "score": 1.426, "rising": -1.31, "first": "2025-11-08", "latest": "2026-07-17", "why": "I use Grok 4 Fast via API, cheap, fast and almost really well suited for data parsing/extraction", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 3, "compare": 0, "warn": 2}, "spark": [1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "I use Grok 4 Fast via API, cheap, fast and almost really well suited for data parsing/extraction", "kind": "", "icon": true, "family": "grok", "domain": "x.ai", "github_repo": null, "official_url": "https://x.ai", "pricing": "", "aliases": ["Grok 4 Fast", "Grok 4.1 Fast"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 768616384, "item_id": 48038524, "url": "https://news.ycombinator.com/item?id=48038524", "by": "petercooper", "date": "2026-05-06", "version": "4.1", "stance": "recommend", "quote": "Grok 4.1 Fast is the cheapest way to do research on X", "excerpt": "I don't know if it relates to the same data centers, but this also comes hours after several still recent Grok models were deprecated at short notice. Grok 4.1 Fast is the cheapest way to do research on X (cheaper than the X API!) and it's gone on May 15: https://docs.x.ai/developers/models - freeing up compute to sell?", "context": "", "use_case": "Research on X", "switched_from": "", "story": "Higher usage limits for Claude and a compute deal with SpaceX"}, {"id": 751490163, "item_id": 46968135, "url": "https://news.ycombinator.com/item?id=46968135", "by": "all2", "date": "2026-02-10", "version": "4.1", "stance": "recommend", "quote": "Both of these choke less than Anthropic's own Sonnet", "excerpt": "My options are 1) pay about a dollar per query from a frontier model, or 2) pay a fraction of that for a not-so-great model that makes my token spend last days/weeks instead of hours. I spend a lot of time on plans, but unfortunately the gotchas are in the weeds, especially when it comes to complex systems. I don't trust these models with even marginally complex, non-standard architectures (my projects center around ", "context": "", "use_case": "", "switched_from": "", "story": "Stop Generating, Start Thinking"}, {"id": 741987648, "item_id": 46374228, "url": "https://news.ycombinator.com/item?id=46374228", "by": "andai", "date": "2025-12-24", "version": "4", "stance": "recommend", "quote": "The one I ended up with was Grok 4 Fast, because it's weirdly fast.", "excerpt": "Thanks. I call this method Power Coding (like Power Armor), where you're still doing everything except for typing out the syntax. I found that for this method the smaller the model, the better it works, because smaller models can generally handle it, and you benefit more from iteration speed than anything else. I don't have hardware to run even tiny LLMs at anything approaching interactive speeds, so I use APIs. The ", "context": "no hardware for local LLMs, uses APIs", "use_case": "", "switched_from": "", "story": "Local AI is driving the biggest change in laptops in decades"}, {"id": 733927776, "item_id": 45870486, "url": "https://news.ycombinator.com/item?id=45870486", "by": "XCSme", "date": "2025-11-09", "version": "4", "stance": "recommend", "quote": "I use Grok 4 Fast via API, cheap, fast and almost really well suited for data parsing/extraction", "excerpt": "I use Grok 4 Fast via API, cheap, fast and almost really well suited for data parsing/extraction, a lot better than Gemini 2.5 Pro for example.", "context": "", "use_case": "data parsing and extraction via API", "switched_from": "", "story": "Grok 4 Fast now has 2M context window"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 783219664, "item_id": 48951229, "url": "https://news.ycombinator.com/item?id=48951229", "by": "Topfi", "date": "2026-07-17", "version": "4", "stance": "warn", "quote": "did severely worse in a similarly outlandish scenario", "excerpt": "Happy to, here one example where Grok 4 Fast, despite producing a fairly consistent pelican [0], did severely worse in a similarly outlandish scenario along with Haiku 4.5 and GPT-5 for context: https://news.ycombinator.com/item?id=45599403 > [...] a really great pelican riding a bicycle and then a terrible sloth riding a skateboard [...] Happy to play ball. You made a blog post a few weeks back on one of the Qwen mo", "context": "", "use_case": "", "switched_from": "", "story": "Kimi K3, and what we can still learn from the pelican benchmark"}, {"id": 775051696, "item_id": 48440731, "url": "https://news.ycombinator.com/item?id=48440731", "by": "andai", "date": "2026-06-07", "version": "4.1", "stance": "warn", "quote": "Requests to grok-4-1-fast-non-reasoning now silently route to grok-4.3 (a 5x more expensive model)", "excerpt": "grok-4-1-fast was retired about a month ago. Requests to grok-4-1-fast-non-reasoning now silently route to grok-4.3 (a 5x more expensive model), with reasoning set to \"none\". https://docs.x.ai/developers/migration/may-15-retirement TFA was published today, which implies grok-4.3 was used.", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeek V4 Pro beats GPT-5.5 Pro on precision"}]}], "who": [{"by": "andai", "context": "no hardware for local LLMs, uses APIs", "url": "https://news.ycombinator.com/item?id=46374228", "date": "2025-12-24"}], "tags": [], "use_cases": ["research on x", "data parsing and extraction via api"], "compared": [{"slug": "openrouter", "name": "OpenRouter", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=45862833", "title": "Grok 4 Fast now has 2M context window", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=48947717", "title": "Kimi K3, and what we can still learn from the pelican benchmark", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48440448", "title": "DeepSeek V4 Pro beats GPT-5.5 Pro on precision", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48037986", "title": "Higher usage limits for Claude and a compute deal with SpaceX", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47078324", "title": "Stop Thinking of AI as a Coworker. It's an Exoskeleton", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46938958", "title": "Stop Generating, Start Thinking", "mentions": 1}], "sentiment_by_month": {"2025-11": {"neutral": 2, "recommend": 1}, "2025-12": {"recommend": 1}, "2026-01": {}, "2026-02": {"recommend": 1, "neutral": 1}, "2026-03": {}, "2026-04": {}, "2026-05": {"recommend": 1}, "2026-06": {"warn": 1}, "2026-07": {"warn": 1}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "family_ref": {"slug": "grok", "name": "Grok"}, "runs_on": null, "lines": [], "quotes": [{"id": 783219664, "item_id": 48951229, "url": "https://news.ycombinator.com/item?id=48951229", "by": "Topfi", "date": "2026-07-17", "version": "4", "stance": "warn", "quote": "did severely worse in a similarly outlandish scenario", "excerpt": "Happy to, here one example where Grok 4 Fast, despite producing a fairly consistent pelican [0], did severely worse in a similarly outlandish scenario along with Haiku 4.5 and GPT-5 for context: https://news.ycombinator.com/item?id=45599403 > [...] a really great pelican riding a bicycle and then a terrible sloth riding a skateboard [...] Happy to play ball. You made a blog post a few weeks back on one of the Qwen mo", "context": "", "use_case": "", "switched_from": "", "story": "Kimi K3, and what we can still learn from the pelican benchmark"}, {"id": 775051696, "item_id": 48440731, "url": "https://news.ycombinator.com/item?id=48440731", "by": "andai", "date": "2026-06-07", "version": "4.1", "stance": "warn", "quote": "Requests to grok-4-1-fast-non-reasoning now silently route to grok-4.3 (a 5x more expensive model)", "excerpt": "grok-4-1-fast was retired about a month ago. Requests to grok-4-1-fast-non-reasoning now silently route to grok-4.3 (a 5x more expensive model), with reasoning set to \"none\". https://docs.x.ai/developers/migration/may-15-retirement TFA was published today, which implies grok-4.3 was used.", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeek V4 Pro beats GPT-5.5 Pro on precision"}, {"id": 768616384, "item_id": 48038524, "url": "https://news.ycombinator.com/item?id=48038524", "by": "petercooper", "date": "2026-05-06", "version": "4.1", "stance": "recommend", "quote": "Grok 4.1 Fast is the cheapest way to do research on X", "excerpt": "I don't know if it relates to the same data centers, but this also comes hours after several still recent Grok models were deprecated at short notice. Grok 4.1 Fast is the cheapest way to do research on X (cheaper than the X API!) and it's gone on May 15: https://docs.x.ai/developers/models - freeing up compute to sell?", "context": "", "use_case": "Research on X", "switched_from": "", "story": "Higher usage limits for Claude and a compute deal with SpaceX"}, {"id": 753293184, "item_id": 47080824, "url": "https://news.ycombinator.com/item?id=47080824", "by": "runarberg", "date": "2026-02-19", "version": "4.1", "stance": "neutral", "quote": "I like this game between grok-4.1-fast and maia-1100", "excerpt": "I like this game between grok-4.1-fast and maia-1100 (engine, not LLM). https://chessbenchllm.onrender.com/game/37d0d260-d63b-4e41-9bba-6b25b922493f This exact game has been played 60 thousand times on lichess. The peace sacrifice Grok performed on move 6 has been played 5 million times on lichess. Every single move Grok made is also the top played move on lichess. This reminds me of Stefan Zweig\u2019s The Royal Game whe", "context": "", "use_case": "", "switched_from": "", "story": "Stop Thinking of AI as a Coworker. It's an Exoskeleton"}, {"id": 751490163, "item_id": 46968135, "url": "https://news.ycombinator.com/item?id=46968135", "by": "all2", "date": "2026-02-10", "version": "4.1", "stance": "recommend", "quote": "Both of these choke less than Anthropic's own Sonnet", "excerpt": "My options are 1) pay about a dollar per query from a frontier model, or 2) pay a fraction of that for a not-so-great model that makes my token spend last days/weeks instead of hours. I spend a lot of time on plans, but unfortunately the gotchas are in the weeds, especially when it comes to complex systems. I don't trust these models with even marginally complex, non-standard architectures (my projects center around ", "context": "", "use_case": "", "switched_from": "", "story": "Stop Generating, Start Thinking"}, {"id": 741987648, "item_id": 46374228, "url": "https://news.ycombinator.com/item?id=46374228", "by": "andai", "date": "2025-12-24", "version": "4", "stance": "recommend", "quote": "The one I ended up with was Grok 4 Fast, because it's weirdly fast.", "excerpt": "Thanks. I call this method Power Coding (like Power Armor), where you're still doing everything except for typing out the syntax. I found that for this method the smaller the model, the better it works, because smaller models can generally handle it, and you benefit more from iteration speed than anything else. I don't have hardware to run even tiny LLMs at anything approaching interactive speeds, so I use APIs. The ", "context": "no hardware for local LLMs, uses APIs", "use_case": "", "switched_from": "", "story": "Local AI is driving the biggest change in laptops in decades"}, {"id": 733927776, "item_id": 45870486, "url": "https://news.ycombinator.com/item?id=45870486", "by": "XCSme", "date": "2025-11-09", "version": "4", "stance": "recommend", "quote": "I use Grok 4 Fast via API, cheap, fast and almost really well suited for data parsing/extraction", "excerpt": "I use Grok 4 Fast via API, cheap, fast and almost really well suited for data parsing/extraction, a lot better than Gemini 2.5 Pro for example.", "context": "", "use_case": "data parsing and extraction via API", "switched_from": "", "story": "Grok 4 Fast now has 2M context window"}, {"id": 733835888, "item_id": 45864743, "url": "https://news.ycombinator.com/item?id=45864743", "by": "mg", "date": "2025-11-09", "version": "4", "stance": "neutral", "quote": "Grok:       Timed out after 300s of thinking", "excerpt": "That makes me wonder if we could simply test this by letting the LLM add or multiply a long list of numbers? Here is an experiment: https://www.gnod.com/search/#q=%23%20Calcuate%20the%20below%20number.%20Do%20not%20use%20a%20calculator.%20Do%20it%20in%20your%20head.%0A%0A(1.1%20%2B%202.5)%20*%20(7.1%20%2B%202.5)%20*%20(3.3%20%2B%204.4)%20*%20(12.3%20-%201.4)%20*%20(2.3%20-%201.1)%20*%20(5%20%2B%201)%20*%20(7.1%20%2B%", "context": "", "use_case": "", "switched_from": "", "story": "Grok 4 Fast now has 2M context window"}, {"id": 733810017, "item_id": 45863126, "url": "https://news.ycombinator.com/item?id=45863126", "by": "minimaxir", "date": "2025-11-08", "version": "4", "stance": "neutral", "quote": "Grok 4 Fast has essentially zero restrictions on safety", "excerpt": "Going off OpenRouter's rankings (https://openrouter.ai/rankings), Grok Code Fast 1 is the most used model by a significant margin, and since those metrics are calculated as of this week, that's after providers stopped giving free promotional access to it. Grok 4 Fast is #5 on that list which was never free. In terms of models, Grok 4 Fast has essentially zero restrictions on safety, which a) makes it unusable for mos", "context": "", "use_case": "", "switched_from": "", "story": "Grok 4 Fast now has 2M context window"}]}