{
stringlengths
1
865
likes: 1,
newest_id: "",
oldest_id: "",
possibly_sensitive: false,
public_metrics: {
BookmarkCount: 0,
ImpressionCount: 0,
LikeCount: 0,
QuoteCount: 0,
ReplyCount: 0,
RetweetCount: 0
},
tweet_id: 1955296671900062000,
user_id: "107463790",
username: "kinopee_ai"
},
created_at: "0001-01-01T00:00:00Z"
},
{
id: "1955296668288471474",
content: "One of the more refreshing behaviors of GPT-5 as a coding model is that not once has it told me, \"Ah, you're absolutely correct!\" 😂",
metadata: {
author: "",
conversation_id: "1955296668288471474",
created_at: "2025-08-12T15:53:48Z",
lang: "",
likes: 0,
newest_id: "",
oldest_id: "",
possibly_sensitive: false,
public_metrics: {
BookmarkCount: 0,
ImpressionCount: 0,
LikeCount: 0,
QuoteCount: 0,
ReplyCount: 0,
RetweetCount: 0
},
tweet_id: 1955296668288471600,
user_id: "6782522",
username: "kgharrison"
},
created_at: "0001-01-01T00:00:00Z"
},
{
id: "1955296592564543957",
content: "AI Model Updates & Chip Tensions\n\n• OpenAI's GPT-5 launched as an iterative update with a unified reasoning system and enhanced multimodal capabilities, receiving mixed user feedback.\n• xAI made Grok 4 free globally, with Grok 4 Heavy remaining premium, claiming performance",
metadata: {
author: "",
conversation_id: "1955296592564543957",
created_at: "2025-08-12T15:53:29Z",
lang: "",
likes: 0,
newest_id: "",
oldest_id: "",
possibly_sensitive: false,
public_metrics: {
BookmarkCount: 0,
ImpressionCount: 0,
LikeCount: 0,
QuoteCount: 0,
ReplyCount: 0,
RetweetCount: 0
},
tweet_id: 1955296592564544000,
user_id: "161923601",
username: "pmainardi"
},
created_at: "0001-01-01T00:00:00Z"
},
{
id: "1955296575720239164",
content: "@elonmusk The percentage that it is “better” than the next closest (GPT-5) is exponentially less than the % that it is both more expensive for inputs and outputs, and % increase in latency.\n\nCope",
metadata: {
author: "",
conversation_id: "1955197567253483672",
created_at: "2025-08-12T15:53:25Z",
lang: "",
likes: 0,
newest_id: "",
oldest_id: "",
possibly_sensitive: false,
public_metrics: {
BookmarkCount: 0,
ImpressionCount: 0,
LikeCount: 0,
QuoteCount: 0,
ReplyCount: 0,
RetweetCount: 0
},
tweet_id: 1955296575720239000,
user_id: "1764104308939710464",
username: "FordGroyper"
},
created_at: "0001-01-01T00:00:00Z"
},
{
id: "1955296563934134389",
content: "@kitsunelynx_ @light_veyne @mark_k @OpenAI Is the API version of GPT-5 Thinking with reasoning set to high smarter than GPT-5 Thinking Pro in the app? I recently tried both and the Pro model gave me by far the better response.",
metadata: {