Hy3 is Tencent's 295B-parameter Mixture-of-Experts model with 21B active parameters, native 256K context, and configurable reasoning modes. It is built for coding, long-context comprehension, multi-turn dialogue, and agentic task execution with a focus on high-throughput production workloads.
Added Jul 6, 2026
Model weightsContext Window
262.1K
Max Output
262.1K
Input Price (Auto)
$0.066/1M
Output Price (Auto)
$0.26/1M
Cache Read (Auto)
$0.029/1M
Capabilities
Benchmarks
Benchmarks
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
25.3
Coding Index
58.8
Agentic work
AutomationBench-AA
Workflow automation with guardrail penalties
18.4%
Better than 42% of models compared
AA-Briefcase
Agentic knowledge work (Elo)
945 Elo
Better than 47% of models compared
GDPval-AA v2
Economically valuable tasks (Elo)
1045 Elo
Better than 54% of models compared
Document reasoning
GDP.pdf
Professional PDF reasoning: all-pass rate
11.2%
Better than 46% of models compared
AA-LCR v1.1
Long context reasoning with updated grading
79.0%
Better than 84% of models compared
Reasoning
HLE
Humanity's Last Exam
33.5%
Better than 83% of models compared
IFBench
Instruction-following benchmark
63.1%
Better than 74% of models compared
Coding
Terminal-Bench v4.0
Practical coding and terminal tasks
0.5%
Better than 37% of models compared
SciCode
Python programming for scientific computing
48.6%
Better than 52% of models compared
Legacy benchmarks
GPQA Diamond (legacy)
Graduate-level scientific reasoning
89.7%
Better than 88% of models compared
Terminal-Bench Hard (legacy)
Agentic coding and terminal use
34.1%
Better than 79% of models compared
T²-Bench Telecom (legacy)
Conversational AI agents in dual-control scenarios
92.7%
Better than 88% of models compared
AA-LCR (unversioned / legacy)
Long context reasoning evaluation
79.0%
Better than 84% of models compared
Last updated Sep 24, 2026
Artificial AnalysisProviders
Provider information for this model’s automatic routing. These routes cannot be selected individually.
Loading provider options…
Related text models
Compare Tencent Hy3 with similar models from the same provider or model family.
Tencent Hy4 Preview
tencent/hy4-previewHy4 Preview is Tencent's 770B-parameter mixture-of-experts model with 49B active parameters. It is designed for coding agents, complex tool-use workflows, and productivity tasks, with a 1M-token context window and configurable reasoning effort.
Ember 1
fireworks/ember-1Fireworks Research reasoning model built on Kimi K3. It accepts text and images, supports tool calling and structured output, and has a 1M-token context window.
Clover 1 150B Preview
kitani/clover-1-150bKitani's 150B multimodal reasoning model for coding, tool use, and everyday assistance. This is a preview release and may have reliability issues.
AionLabs: Aion 3.5
aion-labs/aion-3.5A GLM-family collaborative generation model tuned for immersive roleplay and storytelling, with stronger narrative structure, tension, conflict, and nuanced mature themes.
AionLabs: Aion 3.5 Mini
aion-labs/aion-3.5-miniA GLM-family collaborative generation model tuned for immersive roleplay and storytelling, with stronger narrative structure, tension, conflict, and nuanced mature themes.
Qwen3.8 Max Prime
qwen/qwen3.8-max-primeQwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max for coding, complex analysis, and long-running agent workflows. It accepts text, image, and video input, supports tool calling and structured output, and has a 1M-token context window. Reasoning is always enabled.