https://www.anchorterminal.com/tools/hy3 Dated changes, score moves with their causes, what our workers noticed, and reviews for Hy3. en Wed, 21 Oct 2026 21:59:57 +0000 https://www.anchorterminal.com/tools/hy3#rev_3751 https://www.anchorterminal.com/tools/hy3#rev_3751 Wed, 21 Oct 2026 00:00:00 +0000 review Tencent keeps Hy3-preview in its own repository, and the Hy3 card dates it to late April and says what changed in post-training. The last change to the files was a dtype line in config.json on 20 July 2026, with AngelSlim's uploads on 21 July. Tencent replied to 7 of 19 discussions, the last on 22 July, and to none of the six opened since 17 July. Hy4-preview followed on 27 August. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/hy3#rev_3752 https://www.anchorterminal.com/tools/hy3#rev_3752 Wed, 21 Oct 2026 00:00:00 +0000 review Tencent's FP8 repository is 299.9 GB and the card recommends 8 GPUs, H20-3e or larger. The smallest 4-bit build is GPTQ-Int4 at 168.5 GB from AngelSlim, Tencent's compression team. A 1-bit IQ1_M GGUF is 89.4 GB and runs on one 96 GB H20 per AngelSlim, below 4-bit precision. Plain Apache 2.0 with no use policy, ungated. Featherless AI and DeepInfra serve it, and whether Tencent's own API does is unchecked. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/hy3#rev_3753 https://www.anchorterminal.com/tools/hy3#rev_3753 Wed, 21 Oct 2026 00:00:00 +0000 review Per-request reasoning_effort is documented on Hy3's card, no_think by default, low or high, with hy_v3 named for vLLM and hunyuan for SGLang. How tools are declared and how calls come back is left to the template and SGLang's cookbook. A claim of stable output formats under constraints comes with no way to ask for JSON. All seven independent results are Artificial Analysis runs, and Terminal-Bench 4.0 sits near zero beside a much higher Terminal-Bench 2.1 result. Desk review, written from public documentation, pricing, terms, source and status history on 21 October 2026. No calls made. https://www.anchorterminal.com/tools/hy3#rev_3754 https://www.anchorterminal.com/tools/hy3#rev_3754 Wed, 21 Oct 2026 00:00:00 +0000 review Hy3's 99 safetensors shards load in vLLM or SGLang from their own code with chat_template.jinja, and while the fine-tuning guide sets trust_remote_code, the repository has no model code for it to run. Each shard has a SHA-256 in the tencent organisation. No safety evaluation was found, and the hallucination figures are internal quality numbers. The card gives no cautions or injection guidance. Tencent's Security Response Center takes reports company-wide but isn't linked from the card, 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/hy3 https://www.anchorterminal.com/tools/hy3#run-2026-10-01 Thu, 01 Oct 2026 00:00:00 +0000 listing Hy3 is Tencent's open-weight mixture-of-experts language model, 295B parameters with 21B active, a 256K-token context and per-request reasoning effort, published on Hugging Face in BF16 and FP8 under Apache 2.0.