{"slug": "tencent-hy4-open-weights-model-benchmarks-and-access", "title": "Tencent Hy4 Open-Weights Model: Benchmarks and Access", "summary": "Tencent released Hy4 preview on August 28, a 770B parameter open-weight model under Apache 2.0, with only 49B active parameters per token and a 1 million token context window, targeting software engineering and agentic coding. Hy4 scores 82.9 on SWE-bench Multilingual, ahead of DeepSeek V4 Pro (77.3), Kimi K3 (80.8), and GLM-5.3 (81.3), and is available for free on WorkBuddy and CodeBuddy for two weeks. API pricing via OpenRouter starts at $0.834 per million input tokens, and the FP8 variant requires about 720GB VRAM for self-hosting.", "body_md": "Tencent released Hy4 preview on August 28 — a 770B parameter open-weight model published under Apache 2.0 and available on [Hugging Face](https://huggingface.co/tencent/Hy4-preview) right now. The model uses sparse Mixture-of-Experts routing: only 49B parameters activate per token, keeping inference costs well below what 770B total parameters would suggest. With a 1 million token context window and a built-in speculative decoding layer, it targets software engineering and agentic coding — and Tencent is offering free access on WorkBuddy and CodeBuddy for the next two weeks.\n\nThe competitive positioning is real. Hy4 scores 82.9 on SWE-bench Multilingual, ahead of DeepSeek V4 Pro (77.3), Kimi K3 (80.8), and GLM-5.3 (81.3). API pricing through OpenRouter starts at $0.83 per million input tokens — roughly one-eighth of what GPT-5.6 Sol charges per output token. Apache 2.0 licensing means no field-of-use restrictions and no geographic carve-outs, which is a genuine differentiator in this space.\n\n## How to Access Tencent Hy4 Now\n\nThe fastest path is the free trial on [WorkBuddy and CodeBuddy](https://www.tencent.com/tencent-releases-and-open-sources-tencent-hy4-preview/), Tencent’s developer products. Access runs two weeks from launch, meaning the window closes around September 11. No significant friction beyond creating an account — useful for quick evaluation before committing to API costs.\n\nFor API access, OpenRouter is the launch provider. Pricing: $0.834 per million input tokens, $2.501 per million output tokens. Cached input tokens cost $0.042 per million — a 20x discount that compounds quickly if you’re passing large system prompts or repeated context. Tencent Cloud TokenHub is also available as an alternative endpoint.\n\nSelf-hosting is technically possible under Apache 2.0 but infrastructure-intensive. The FP8 quantized variant requires approximately 720GB of VRAM — achievable on a single 8xH200 node, but beyond what 8xH100 nodes offer (640GB max). Prebuilt Docker images exist for both vLLM and SGLang. For most teams, OpenRouter is the practical starting point.\n\n```\ndocker run --gpus all -p 8000:8000 --ipc=host \\\n  vllm/vllm-openai:hy4-preview tencent/Hy4-preview-FP8 \\\n  --tensor-parallel-size 8 --served-model-name hy4-preview\n```\n\n## Tencent Hy4 Benchmarks: Where It Wins and Loses\n\nHy4 leads open-weight peers where multi-language code matters. Its 82.9 on SWE-bench Multilingual beats DeepSeek V4 Pro (77.3), Kimi K3 (80.8), and GLM-5.3 (81.3). However, Terminal-Bench 2.1 tells a different story: Hy4 scores 85.4 while GLM-5.3 hits 88.2 and Kimi K3 reaches 88.3. For agentic coding tasks requiring sustained tool use, it’s competitive but not top-tier.\n\nThe generational jump from Hy3 is substantial. DeepSWE — a measure of real-world software engineering — climbed from 28.0 on Hy3 to 64.3 on Hy4, a 2.3x improvement in one model cycle. Claude Opus 5 still leads on SWE-bench Pro (79.2 vs Hy4’s 65.7), so if maximum coding accuracy is the requirement, the gap remains real. One important caveat: all Hy4 benchmark scores come from [Tencent’s internal evaluators](https://technode.com/2026/08/28/tencent-open-sources-hy4-preview-with-770b-parameters-and-a-1m-token-context/) on proprietary task sets. No independent leaderboard verification exists yet.\n\n## The Preview Caveats Worth Knowing\n\nTencent labels this an early version, and the limitations are real. Hy4 defaults to deep chain-of-thought reasoning and tends to over-verify its own work on straightforward tasks — inflating latency where none is needed. The fix is a single parameter change when calling the API:\n\n```\nresponse = client.chat.completions.create(\n    model=\"tencent/hy4-preview\",\n    messages=[{\"role\": \"user\", \"content\": \"Review this function for bugs\"}],\n    extra_body={\"chat_template_kwargs\": {\"reasoning_effort\": \"no_think\"}}\n)\n```\n\nAdditionally, [OpenRouter measured roughly 86% availability](https://explainx.ai/blog/tencent-hy4-preview-770b-moe-1m-context-august-2026) in the model’s first three days — still stabilizing after launch. Tencent acknowledges headroom remains in both pre-training and post-training, meaning a more capable final version follows this preview. For production workloads with uptime requirements, wait for the stable release.\n\n## Key Takeaways\n\n- Hy4 preview is live now: free on WorkBuddy and CodeBuddy until around September 11, then $0.83 per million input tokens via OpenRouter\n- It leads open-weight peers on SWE-bench Multilingual (82.9) but trails on Terminal-Bench 2.1 — strong for multi-language code review, middle tier for agentic tasks\n- Apache 2.0 licensing is genuinely unrestricted — commercial use, modifications, and self-hosting all permitted without carve-outs\n- Set\n`reasoning_effort: \"no_think\"`\n\nfor direct responses; the default deep reasoning mode adds unnecessary latency on simple tasks - Self-hosting needs 8xH200 GPUs minimum for the FP8 variant — most teams will use OpenRouter until infrastructure scales", "url": "https://wpnews.pro/news/tencent-hy4-open-weights-model-benchmarks-and-access", "canonical_source": "https://byteiota.com/tencent-hy4-open-weights-model-benchmarks-and-access/", "published_at": "2026-08-30 08:11:00+00:00", "updated_at": "2026-08-30 08:21:55.596746+00:00", "lang": "en", "topics": ["large-language-models", "ai-products", "ai-tools", "ai-research"], "entities": ["Tencent", "Hy4", "DeepSeek V4 Pro", "Kimi K3", "GLM-5.3", "OpenRouter", "WorkBuddy", "CodeBuddy"], "alternates": {"html": "https://wpnews.pro/news/tencent-hy4-open-weights-model-benchmarks-and-access", "markdown": "https://wpnews.pro/news/tencent-hy4-open-weights-model-benchmarks-and-access.md", "text": "https://wpnews.pro/news/tencent-hy4-open-weights-model-benchmarks-and-access.txt", "jsonld": "https://wpnews.pro/news/tencent-hy4-open-weights-model-benchmarks-and-access.jsonld"}}