{"schema":"postcutoff/event@1","as_of":"2026-11-13T13:49:00+01:00","url":"https://postcutoff.com/e/2024-12-26-deepseek-v3/","md":"https://postcutoff.com/e/2024-12-26-deepseek-v3/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2024-12-26-deepseek-v3","date":"2024-12-26","date_precision":"day","short_title":"DeepSeek-V3: frontier-level open model trained for ~$5.6M in GPU time","deck":null,"takeaway":"Chinese lab DeepSeek released DeepSeek-V3, a 671B-parameter mixture-of-experts model (37B active) with open weights that rivaled GPT-4o and Claude 3.5 Sonnet; its final training run reportedly used 2.788M H800 GPU-hours (~$5.6M).","category":"open-source","category_label":"Open source","importance":5,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"DeepSeek-V3 Technical Report (arXiv)","url":"https://arxiv.org/abs/2412.19437","type":"paper","group":"primary","domain":"arxiv.org"},{"n":2,"title":"deepseek-ai/DeepSeek-V3 (code & weights)","url":"https://github.com/deepseek-ai/DeepSeek-V3","type":"code","group":"primary","domain":"github.com"}],"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["DeepSeek"],"title":"DeepSeek-V3: frontier-level open model trained for ~$5.6M in GPU time","summary":"Chinese lab DeepSeek released DeepSeek-V3, a 671B-parameter mixture-of-experts model (37B active) with open weights that rivaled GPT-4o and Claude 3.5 Sonnet; its final training run reportedly used 2.788M H800 GPU-hours (~$5.6M).","key_facts":["Released 26 December 2024; technical report arXiv 2412.19437","671B total parameters, 37B activated per token","Pre-trained on 14.8 trillion tokens","2.788M H800 GPU-hours for full training (~$5.576M at $2/GPU-hour, excluding prior research)","Innovations: multi-head latent attention, auxiliary-loss-free load balancing, FP8 training, multi-token prediction"],"key_numbers":[],"tags":["llm","open-weights","mixture-of-experts","china","efficiency"],"science":null,"body_md":"## What happened\n\nDeepSeek published open weights and an unusually detailed report showing frontier performance at a fraction of the reported compute of US labs, despite export controls.\n\n## Why it matters\n\nUpended assumptions about the cost of frontier AI and China's position; it was the base for DeepSeek-R1 weeks later.","disputed":[],"related":[{"id":"2025-01-20-deepseek-r1","url":"https://postcutoff.com/e/2025-01-20-deepseek-r1/","date":"2025-01-20","date_precision":"day","short_title":"DeepSeek-R1: open-weights reasoning model rivals o1 and shakes markets","deck":null,"takeaway":"The 'DeepSeek moment' showed that frontier reasoning could be replicated cheaply and openly, triggering a market shock, a wave of open reasoning models, and US policy debates on China.","category":"open-source","category_label":"Open source","importance":5,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":3,"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["DeepSeek"]},{"id":"2023-12-11-mixtral","url":"https://postcutoff.com/e/2023-12-11-mixtral/","date":"2023-12-11","date_precision":"day","short_title":"Mistral AI releases Mixtral 8x7B, an open mixture-of-experts model","deck":null,"takeaway":"Popularized mixture-of-experts in open models (later used by DeepSeek-V3, Llama 4, Qwen) and established Europe's leading AI startup.","category":"open-source","category_label":"Open source","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":2,"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Mistral AI"]}],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-09-29","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-09-29","run":null,"sources_read":null,"updated":"2026-09-29","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":null,"in_training_data":true},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":null,"in_training_data":true},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":null,"in_training_data":true},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":null,"in_training_data":true}],"short_url":"https://postcutoff.com/s/deepseek-v3-frontier-level-open"}