LLMtencent

Tencent: Hy3

Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...

Anyone in the Project can @-mention Tencent: Hy3 with the team's shared context - pooled credits, one chat, one memory.

All models

Starter is free forever - 1 Project, 100 credits/month, 1 MCP. No card.

Verdict

Tencent's Hy3 delivers a massive 202K token context window at $0.14/$0.58 per Mtok — roughly half the cost of GPT-4o for long-document work. Without public benchmarks, you're trading proven performance data for price and capacity. Best for teams already in Tencent's ecosystem who need affordable long-context inference and can validate quality in-house before production use.

Best for

  • Long-document analysis on tight budgets
  • Multi-document synthesis under 200K tokens
  • Cost-sensitive chatbot backends
  • Tencent Cloud native integrations

Strengths

The 202K context window handles full codebases, legal briefs, or multi-chapter manuscripts in a single call. At $0.14 input per Mtok, it undercuts most frontier models on long-context tasks where you're feeding large volumes of text. Tencent's infrastructure means low-latency access for teams in Asia-Pacific, and the proprietary license suggests active support for enterprise customers.

Trade-offs

No public benchmarks means you cannot compare reasoning quality, instruction-following, or factual accuracy against Claude, GPT-4, or Gemini before committing. The model is new enough that community tooling, fine-tuning recipes, and third-party evaluations are sparse. Output quality on complex reasoning or creative tasks remains unvalidated outside Tencent's own claims.

Specifications

Provider
tencent
Category
llm
Context length
202,752 tokens
Max output
131,072 tokens
Modalities
text
License
proprietary
Released
2026-07-06

Pricing

Input
$0.14/Mtok
Output
$0.58/Mtok
Model ID
tencent/hy3

Per-token prices show what the model costs upstream. On Switchy your team draws from one shared org credit pool - one plan, one balance for everyone.

Team cost calculator

Estimated monthly spend
$4.79
17.6M tokens / month
5 seats · 80 msgs/day

Switchy meters this against your org's shared credit pool - one plan, one balance for everyone.

Providers

Provider-level routing data is not available yet for this model.

Performance

Performance snapshots are collected daily. Check back after the next ingestion run.

Benchmarks

Public benchmark scores are not available yet for this model. Check back after the next ingestion run.

Works well with

Top MCPs

Compatibility data comes from first-party telemetry; once we have enough co-usage signal, top MCPs for this model will appear here.

How Switchy teams use it

Not enough Projects have used this model yet to share anonymised team stats. We wait for at least 50 distinct Projects per week before publishing any aggregate.

Starter prompts

Compare Research Papers

I've pasted three research papers below. Compare their methodologies, highlight where findings agree or conflict, and note any gaps in the literature they collectively reveal.
Open in a Project →

Analyze Codebase Structure

Here's the full source tree for a Python web app. List the main modules, describe the data flow from HTTP request to database write, and flag any unused imports or circular dependencies.
Open in a Project →

Extract Meeting Insights

Below is a transcript of a three-hour planning meeting. Extract all action items with owners, decisions made, and questions that remain open. Group by topic.
Open in a Project →

Draft Multi-Chapter Outline

I've provided background research and a thesis statement. Draft a chapter-by-chapter outline for a non-fiction book, with 2-3 bullet points per chapter describing key arguments and evidence.
Open in a Project →
Data last verified 6 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.