Hy3 Official Specifications, Pricing & Benchmarks
Hy3 (hy3) by Tencent 腾讯混元, released 2026-07-06, featuring 262,144 context window, 128,000 max output tokens, and Text input.
📋 Hy3 Official API & System Card Specifications
Verified against official Tencent 腾讯混元 developer documentation & pricing
1. Identity & Capabilities
API Spec- API Model ID
- hy3
- Release Date
- 2026-07-06
- Weight License
- Open-Weight (API Spec)
- Native Reasoning
- ✅ Supported
- Function Calling & JSON Schema
- ✅ 原生支持
- Batch API 异步批处理
- ✅ Supported (50% off)
2. Context, Output & Modalities
Limits- Max Context Window
- 262,000 tokens (262K)
- Max Output Tokens
- 128,000 tokens
- Input Modalities
- Text
- Output Modalities
- Text
- Direct China Access
- ✅ Direct CN + RMB
- Deployment Mode
- Open-Weight + Cloud API
3. Official Pricing & Cache
USD / 1M- Input / 1M Tokens
- $ 0.132
- Cached Input / 1M
- $ 0.033 -75%
- Output / 1M Tokens
- $ 0.528
🎛️ Official Sampling & Reasoning Parameters
Recommended generation settings & API parameter constraints
📊 Hy3 Full Benchmark Scorecard
29 public benchmarks · ranks among 154 models · synced 2026-11-20
💻 Coding 11 items
MCP Atlas is a benchmark for evaluating AI models on scaled tool use capabilities, measuring how well models can coordinate and utilize multiple tools across complex multi-step tasks.
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
A multilingual benchmark for issue resolving in software engineering that covers Java, TypeScript, JavaScript, Go, Rust, C, and C++.
Terminal-Bench 2.1 is an updated release of the Terminal-Bench benchmark that tests AI agents' ability to operate a computer via the terminal.
Claw-Eval tests real-world agentic task completion across complex multi-step scenarios, evaluating a model's ability to use tools, navigate environments, and complete end-to-end tasks autonomously.
SWE-Bench Pro is an advanced version of SWE-Bench that evaluates language models on complex, real-world software engineering tasks requiring extended reasoning and multi-step problem solving.
SkillsBench evaluates coding agents on self-contained programming tasks, measuring practical engineering skills across diverse software development scenarios.
WildClawBench is an agentic coding benchmark from InternLM/Claw-Eval that reports overall model performance on real-world tool-using development tasks.
NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.
DeepSWE is a software engineering agent benchmark evaluated with the mini-swe-agent harness, where each task is solved in an isolated container with no internet access. It measures an agent's ability to autonomously resolve real-world coding issues end to end.
CL-bench is an open-source benchmark with its own data and rubrics for evaluating models on coding and agentic tasks, scored using a setup fully aligned with the official procedure.
🤖 Agents & Tools 5 items
DeepSearchQA is a benchmark for evaluating deep search and question-answering capabilities, testing models' ability to perform multi-hop reasoning and information retrieval across complex knowledge domains.
BrowseComp is a benchmark comprising 1,266 questions that challenge AI agents to persistently navigate the internet in search of hard-to-find, entangled information.
WideSearch is an agentic search benchmark that evaluates models' ability to perform broad, parallel search operations across multiple sources. It tests wide-coverage information retrieval and synthesis capabilities.
Tool Decathlon is a comprehensive benchmark for evaluating AI agents' ability to use multiple tools across diverse task categories. It measures proficiency in tool selection, sequencing, and execution across ten different tool-use scenarios.
APEX-Agents is a benchmark evaluating AI agents on long horizon professional tasks that require sustained reasoning, planning, and execution across complex multi-step workflows.
📐 Math 8 items
IMO-AnswerBench is a benchmark for evaluating mathematical reasoning capabilities on International Mathematical Olympiad (IMO) problems, focusing on answer generation and verification.
FrontierScience Olympiad is a benchmark of olympiad-level scientific reasoning problems, testing expert understanding and multi-step reasoning across advanced natural-science domains.
USAMO 2026 evaluates models on the six problems from the 2026 United States of America Mathematical Olympiad, requiring rigorous proof-based reasoning.
Text-only Humanity's Last Exam variant evaluated with tool use enabled.
ArXivMath is a final-answer benchmark of research-level mathematics maintained by MathArena. Problems are extracted monthly from recent arXiv paper abstracts, then filtered through automated and manual checks to ensure they are self-contained, non-trivial, and verifiable.
Text-only Humanity's Last Exam variant evaluated without tool use.
MathArena Apex is a challenging math contest benchmark featuring the most difficult mathematical problems designed to test advanced reasoning and problem-solving abilities of AI models. It focuses on olympiad-level mathematics and complex multi-step mathematical reasoning.
HorizonMath is an extremely difficult frontier mathematics benchmark designed to test the limits of mathematical reasoning on research-level and competition-beyond problems.
📜 Long Context 1 items
🧠 Reasoning & Science 4 items
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
PHYBench is a benchmark of real-world physics problems spanning mechanics, electromagnetism, thermodynamics, optics, and modern physics, designed to evaluate physical perception and multi-step quantitative reasoning in large language models.
SuperChem is a benchmark of advanced chemistry problems requiring expert-level domain knowledge and reasoning.
FrontierScience Research is a benchmark evaluating AI models on cutting-edge scientific research questions requiring deep domain expertise, multi-step reasoning, and synthesis of complex scientific concepts across disciplines.
Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).
Closest Self-Hostable Open-Weight Alternatives to Hy3
Open-weight models with full HuggingFace weights for private GPU / Mac deployment
Frequently Asked Questions
Is Hy3 open-source or proprietary?
Hy3 (hy3) is a proprietary commercial model by Tencent 腾讯混元. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.
What are the context limit and API pricing of Hy3?
Hy3 supports 262K input context and 128,000 max output tokens, priced at $0.132/1M input ($0.033/1M cached) and $0.528/1M output.