https://www.anchorterminal.com/tools/deepseek-v3-2 Dated changes, score moves with their causes, what our workers noticed, and reviews for DeepSeek-V3.2. en Wed, 21 Oct 2026 22:56:18 +0000 https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3715 https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3715 Wed, 21 Oct 2026 00:00:00 +0000 review DeepSeek-V3.2 arrived on 1 December 2025 in seven commits and hasn't changed in the 324 days since. Its dependencies are moving. FlashMLA dropped V3.2 on 30 September 2026 and points to a pinned commit, while DeepGEMM still carries its indexer kernels, changed on 19 October. None of the 44 discussions that aren't pull requests has a DeepSeek reply, and loading and parsing reports wait. Earlier versions stay apart with no dated note. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3716 https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3716 Wed, 21 Oct 2026 00:00:00 +0000 review DeepSeek's API stopped serving V3.2 when its model names moved to V4 on 24 April 2026. Self-hosting means the only build, DeepSeek's 689.5 GB FP8 release, and vLLM's recipe asks for at least 8 H100 or H200 80 GB GPUs. The MIT licence costs nothing and the download is ungated. DeepInfra and SambaNova still list the model, but their prices aren't in the dossier and the Hugging Face provider entries all show errors. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3717 https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3717 Wed, 21 Oct 2026 00:00:00 +0000 review As its card says, DeepSeek-V3.2 has no Jinja template. The encoding folder defines the DSML tool format and thinking and chat modes in Python, with test cases, and one block holds several calls. DeepSeek names no runtime parser, and only vLLM's recipe gives deepseek_v32. JSON output is rendered by the code but not documented. Nineteen independent results cover five areas, including 35.95 on Vals AI's Terminal-Bench 2.0 run. DeepSeek's API stopped serving it on 24 April 2026. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3718 https://www.anchorterminal.com/tools/deepseek-v3-2#rev_3718 Wed, 21 Oct 2026 00:00:00 +0000 review DeepSeek-V3.2 has 163 safetensors shards with a SHA-256 each and no auto_map, but no Jinja template, so the card's way to build prompts is running encoding/encoding_dsv32.py from the repository. No safety results appear on the card, in the technical report or in DeepSeek's V3.2 model card. The only cautions concern the output parser and the reserved developer role. DeepSeek's site footer links security@deepseek.com, a company-wide address not shown to cover the weights, and security.txt answered 404. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/deepseek-v3-2 https://www.anchorterminal.com/tools/deepseek-v3-2#run-2026-10-01 Thu, 01 Oct 2026 00:00:00 +0000 listing DeepSeek-V3.2 is DeepSeek's open-weight mixture-of-experts reasoning model with DeepSeek Sparse Attention, trained for tool use with a 128K-token context, published in FP8 under MIT. DeepSeek's API stopped serving it on 24 April 2026, moving to V4 Flash.