{"schema_version":"onlylabs.public_signal.v1","title":"Baseten Writing: 22580 Gpt 2 To Kimi K3 Explained","description":"Baseten writing signal with public source context, captured evidence pages, related signals, and category-scoped analysis context.","url":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7","json_url":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7/signal.json","generated_at":"2026-09-10T00:03:12.214Z","evidence_latest_fetched_at":"2026-07-30T16:02:47.90867+00:00","signal_first_seen_at":"2026-07-30T16:01:43.426377+00:00","org":{"slug":"baseten","name":"Baseten","category":"neocloud","category_label":"Neocloud","dossier_url":"https://onlylabs.fyi/labs/baseten","dossier_json_url":"https://onlylabs.fyi/labs/baseten/dossier.json"},"related_urls":{"signal":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7","signal_json":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7/signal.json","source":"https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/","lab_dossier":"https://onlylabs.fyi/labs/baseten","lab_dossier_json":"https://onlylabs.fyi/labs/baseten/dossier.json","analysis":"https://onlylabs.fyi/analysis/baseten","analysis_json":"https://onlylabs.fyi/analysis/baseten/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/baseten/evidence.json","category":"https://onlylabs.fyi/neoclouds","category_json":"https://onlylabs.fyi/neoclouds.json","category_feed":"https://onlylabs.fyi/neoclouds/feed.xml","category_signals_json":"https://onlylabs.fyi/signals.json?category=neocloud","topic":"https://onlylabs.fyi/topics/talking","topic_signals_json":"https://onlylabs.fyi/topics/talking/signals.json?category=neocloud","topic_feed":"https://onlylabs.fyi/topics/talking/feed.xml?category=neocloud","data_business":null},"answer_pack":{"answer":"Baseten published 22580 Gpt 2 To Kimi K3 Explained. This talking signal gives public context for research themes, product direction, policy, or launch framing. High-signal details: Explainer post on model evolution. · 22,580: GPT-2 to Kimi K3, explained Kimi K3 is here. Try it now Foundations 22,580: GPT-2 to Kimi K3, explained Twenty-two thousand five hundred and eighty. That’s how.... onlylabs links this event to 1 captured evidence page and 6 related writing signals.","signal_desk":"talking","source_context":{"source_url":"https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/","source_host":"baseten.co","occurred_at":"2026-07-30T00:00:00.000Z","first_seen_at":"2026-07-30T16:01:43.426377+00:00","date_source":"page.visible_date","context":null},"context_markers":[{"label":"Lab","value":"Baseten","source":"signal"},{"label":"Signal desk","value":"talking","source":"signal"},{"label":"Source host","value":"baseten.co","source":"source"},{"label":"Notability","value":"Explainer post on model evolution.","source":"signal"},{"label":"Watch term","value":"Infrastructure","source":"evidence"}],"evidence_coverage":{"target_pages":1,"captured_pages":1,"readable_pages":1,"capture_methods":["plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/"],"related_signals":6,"has_source_url":true,"latest_page_fetched_at":"2026-07-30T16:02:47.90867+00:00"},"data_business":{"matches":false,"lanes":[],"matched_terms":[],"score":null,"reason":null},"agent_handoff":{"signal_json":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7/signal.json","dossier_json":"https://onlylabs.fyi/labs/baseten/dossier.json","analysis_json":"https://onlylabs.fyi/analysis/baseten/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/baseten/evidence.json","topic_signals_json":"https://onlylabs.fyi/topics/talking/signals.json?category=neocloud","topic_feed":"https://onlylabs.fyi/topics/talking/feed.xml?category=neocloud","category_signals_json":"https://onlylabs.fyi/signals.json?category=neocloud","data_radar_json":null,"opportunities_json":null},"analysis_playbook":{"objective":"Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.","evidence_focus":["post title","source URL","captured page text","HN traction","linked model or paper references","publication date"],"extraction_questions":["Which themes are labs choosing to explain publicly?","Which posts are attracting outside discussion?","Which writing reframes a recent release, model, hiring wave, or policy stance?","Which posts mention data, evals, infrastructure, safety, or deployment workflows?"],"signal_questions":["What public theme, launch framing, or research direction does this writing signal expose?","Which themes are labs choosing to explain publicly?","Which posts are attracting outside discussion?","Do the 6 related writing signals show a repeated pattern?"],"output_fields":["org","theme","public_framing","traction","evidence_url"],"data_business_relevance":"Data-business lane extraction is scoped to frontier labs; for this category, keep conclusions tied to category-specific strategy, source evidence, and follow-up questions.","required_sources":[{"label":"signal_json","url":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7/signal.json","required":true},{"label":"source","url":"https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/","required":true},{"label":"dossier_json","url":"https://onlylabs.fyi/labs/baseten/dossier.json","required":true},{"label":"analysis_evidence_json","url":"https://onlylabs.fyi/analysis/baseten/evidence.json","required":true},{"label":"topic_signals_json","url":"https://onlylabs.fyi/topics/talking/signals.json?category=neocloud","required":false},{"label":"data_radar_json","url":null,"required":false}],"expected_output":["one-paragraph source-grounded interpretation","category-specific implication","confidence and missing evidence","recommended next source to inspect"],"prompt_seed":"Using only the linked onlylabs JSON, captured source context, and cited evidence, analyze Baseten's writing signal \"22580 Gpt 2 To Kimi K3 Explained\" for neocloud strategy."},"semantic_triples":[{"subject":"Baseten","predicate":"published","object":"22580 Gpt 2 To Kimi K3 Explained","text":"Baseten published 22580 Gpt 2 To Kimi K3 Explained."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"is classified as","object":"writing signal","text":"22580 Gpt 2 To Kimi K3 Explained is classified as writing signal."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"belongs to","object":"talking desk","text":"22580 Gpt 2 To Kimi K3 Explained belongs to talking desk."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has evidence coverage","object":"1 captured evidence page","text":"22580 Gpt 2 To Kimi K3 Explained has evidence coverage 1 captured evidence page."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has captured page count","object":"1","text":"22580 Gpt 2 To Kimi K3 Explained has captured page count 1."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has readable page count","object":"1","text":"22580 Gpt 2 To Kimi K3 Explained has readable page count 1."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has related signal count","object":"6","text":"22580 Gpt 2 To Kimi K3 Explained has related signal count 6."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has analysis playbook objective","object":"Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.","text":"22580 Gpt 2 To Kimi K3 Explained has analysis playbook objective Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has source host","object":"baseten.co","text":"22580 Gpt 2 To Kimi K3 Explained has source host baseten.co."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has lab","object":"Baseten","text":"22580 Gpt 2 To Kimi K3 Explained has lab Baseten."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has signal desk","object":"talking","text":"22580 Gpt 2 To Kimi K3 Explained has signal desk talking."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has source host","object":"baseten.co","text":"22580 Gpt 2 To Kimi K3 Explained has source host baseten.co."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has notability","object":"Explainer post on model evolution.","text":"22580 Gpt 2 To Kimi K3 Explained has notability Explainer post on model evolution.."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has watch term","object":"Infrastructure","text":"22580 Gpt 2 To Kimi K3 Explained has watch term Infrastructure."}]},"intelligence":{"signal_desk":"talking","answer":"Baseten published 22580 Gpt 2 To Kimi K3 Explained. This talking signal gives public context for research themes, product direction, policy, or launch framing. High-signal details: Explainer post on model evolution. · 22,580: GPT-2 to Kimi K3, explained Kimi K3 is here. Try it now Foundations 22,580: GPT-2 to Kimi K3, explained Twenty-two thousand five hundred and eighty. That’s how.... onlylabs links this event to 1 captured evidence page and 6 related writing signals.","semantic_triples":[{"subject":"Baseten","predicate":"published","object":"22580 Gpt 2 To Kimi K3 Explained","text":"Baseten published 22580 Gpt 2 To Kimi K3 Explained."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"is classified as","object":"writing signal","text":"22580 Gpt 2 To Kimi K3 Explained is classified as writing signal."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"belongs to","object":"talking desk","text":"22580 Gpt 2 To Kimi K3 Explained belongs to talking desk."},{"subject":"22580 Gpt 2 To Kimi K3 Explained","predicate":"has evidence coverage","object":"1 captured evidence page","text":"22580 Gpt 2 To Kimi K3 Explained has evidence coverage 1 captured evidence page."}]},"signal":{"id":"1e742e9b-ddc2-4855-8458-4cb6d6fad1e7","url":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7","json_url":"https://onlylabs.fyi/signals/1e742e9b-ddc2-4855-8458-4cb6d6fad1e7/signal.json","source_url":"https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/","title":"22580 Gpt 2 To Kimi K3 Explained","summary":"Baseten published a writing signal. onlylabs watches public writing for research themes, product direction, and model-launch context.","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-30T00:00:00.000Z","first_seen_at":"2026-07-30T16:01:43.426377+00:00","date_source":"page.visible_date","evidence_coverage":{"target_pages":1,"captured_pages":1,"readable_pages":1,"capture_methods":["plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/"]},"facets":{},"traction":{"github_stars":null,"hn_points":null,"hn_comments":null,"hn_story_id":null,"hf_downloads":null,"hf_likes":null},"data_radar":null},"primary_evidence_page":{"is_primary":true,"source_match":true,"url":"https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/","final_url":"https://www.baseten.co/blog/22580-gpt-2-to-kimi-k3-explained/","title":"22580 Gpt 2 To Kimi K3 Explained","http_status":200,"content_type":"text/html; charset=utf-8","capture_method":"plain","fetched_at":"2026-07-30T16:02:47.90867+00:00","bytes":577814,"raw_path":"8fbaf5775830f51c1f0c120d420adc06939f193e240af2a06152ab36f0e421b4.html","content_hash":"d0b9eed5d199d5364ae6bf019ccd9415c6e9a121b82ade666fb259f70142ac16","excerpt_chars":1200,"truncated":true,"excerpt":"22,580: GPT-2 to Kimi K3, explained Kimi K3 is here. Try it now Foundations 22,580: GPT-2 to Kimi K3, explained Twenty-two thousand five hundred and eighty. That’s how many GPT-2 (2019) models fit inside KimiK3 (2026). Authors Ali Taha Last updated July 30, 2026 Share TL;DR Twenty-two thousand five hundred and eighty. That’s how many GPT-2 (2019) models fit inside Kimi K3 (2026). We scaled up by a factor of 22,580 in seven years. But is it just... scale? In this worklog, I’ll walk through how we got here and how much, or how little, has actually changed since then. We’ll trace the major architectural developments leading to Kimi K3. ✕ GPT-2 GPT-2 is a decoder-only architecture: tok_emb = self.transformer.wte(idx) # token embeddings of shape (b, t, n_embd) pos_emb = self.transformer.wpe(pos) # position embeddings of shape (t, n_embd) x = self.transformer.drop(tok_emb + pos_emb) for block in self.transformer.h: x = block(x) x = self.transformer.ln_f(x) logits = self.lm_head(x) return logits The input receives token and positional embeddings: ✕ Each transformer block, zoomed in, looks like this: 1 class Block ( nn.Module ): 2 def __init__ ( self, config ): 3 super ().__init__() 4..."},"evidence_pages":[],"related_signals":[{"id":"7b42fac2-f048-41b3-9275-349c47f2335a","url":"https://onlylabs.fyi/signals/7b42fac2-f048-41b3-9275-349c47f2335a","source_url":"https://www.baseten.co/blog/how-baseten-makes-pyannotes-diarization-models-96x-faster/","title":"How Baseten Makes Pyannotes Diarization Models 96x Faster","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-09-09T00:00:00+00:00","first_seen_at":"2026-09-09T20:00:53.077479+00:00","date_source":"sitemap.lastmod"},{"id":"5a040dc3-5085-482d-bb70-4e9096d0e574","url":"https://onlylabs.fyi/signals/5a040dc3-5085-482d-bb70-4e9096d0e574","source_url":"https://www.baseten.co/blog/optimizing-delta-weight-syncs-for-managed-rollouts/","title":"Optimizing Delta Weight Syncs For Managed Rollouts","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-09-08T22:36:24+00:00","first_seen_at":"2026-09-09T00:00:51.730848+00:00","date_source":"sitemap.lastmod"},{"id":"4bc6b810-0677-4409-a1bb-b8d733ecb37b","url":"https://onlylabs.fyi/signals/4bc6b810-0677-4409-a1bb-b8d733ecb37b","source_url":"https://www.baseten.co/blog/baseten-leads-covals-voice-ai-benchmark/","title":"Baseten Leads Covals Voice Ai Benchmark","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-09-08T22:20:40+00:00","first_seen_at":"2026-09-09T00:00:51.730848+00:00","date_source":"sitemap.lastmod"},{"id":"3af4f3d3-59a4-4686-94ba-6b6cacdd1091","url":"https://onlylabs.fyi/signals/3af4f3d3-59a4-4686-94ba-6b6cacdd1091","source_url":"https://www.baseten.co/blog/new-mcp-and-skill-for-coding-agents-to-use-baseten/","title":"New Mcp And Skill For Coding Agents To Use Baseten","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-09-02T18:20:19+00:00","first_seen_at":"2026-09-02T20:00:52.042916+00:00","date_source":"sitemap.lastmod"},{"id":"8dd84285-f6fb-49cb-a423-7039fcdde208","url":"https://onlylabs.fyi/signals/8dd84285-f6fb-49cb-a423-7039fcdde208","source_url":"https://www.baseten.co/blog/best-open-source-models-for-post-training/","title":"Best Open Source Models For Post Training","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-09-02T10:51:41+00:00","first_seen_at":"2026-09-02T12:01:45.497094+00:00","date_source":"sitemap.lastmod"},{"id":"9848b185-9578-4f96-8911-a25dba91ec70","url":"https://onlylabs.fyi/signals/9848b185-9578-4f96-8911-a25dba91ec70","source_url":"https://www.baseten.co/blog/the-efficient-frontier-of-llm-inference/","title":"The Efficient Frontier Of Llm Inference","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-09-01T23:47:31+00:00","first_seen_at":"2026-09-02T00:00:51.666285+00:00","date_source":"sitemap.lastmod"}]}