Hy3 pricing, context and benchmark
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.
Everything that matters, in one table
Context, pricing, cache, modalities and tooling, from the model registry and Artificial Analysis.
| Model | Hy3 |
| Model ID | tencent/hy3 |
| Provider | Tencent |
| Release date | 6 July 2026 |
| Status | Generally available |
| Input context | 262,144 tokens |
| Output context | 235,929 tokens |
| Official input pricing | $0.132 per 1M tokens |
| Official output pricing | $0.528 per 1M tokens |
| Cache read pricing | $0.035 per 1M tokens (75% off input) |
| Market input pricing | $0.083 per 1M (cheapest route) |
| Market output pricing | $0.33 per 1M (cheapest route) |
| Highest route pricing | $0.2 in, $0.8 out per 1M |
| Input modalities | text |
| Output modalities | text |
| Reasoning control | Not exposed |
| Tool support | Text generation only |
| Tokenizer | Other |
| Provider routes | 5 endpoints |
| Artificial Analysis index | Not published |
| Speed | 64 tokens/s, 41.28s to first token |
| Knowledge cutoff | Not publicly stated |
Input and output support
| Modality | Input | Output | Billed at |
|---|---|---|---|
| Text | ✓ yes | ✓ yes | $0.132 in |
| Image | no | no | not supported |
| Audio | no | no | not supported |
| Video | no | no | not supported |
| File (PDF) | no | no | not supported |
What is 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:...
Hy3 carries a 262K token window, uses the Other tokenizer and is served by Tencent. On Hoodgen it sits in the Reasoning tier of the model menu next to its siblings, so you can switch mid conversation without losing the thread.
Release and what changed from the previous version
This is the only release of Hy3 on OpenRouter, released 6 July 2026. The provider has not published a knowledge cutoff, so treat training data as recent but unconfirmed for very new events.
Features and capabilities
What can you do with Hy3?
Prompt caching and how to cut the bill
Cached input bills at $0.035 per million instead of $0.132, a 75% discount. Keep the system prompt and documents first and the changing question last, so every call hits the cache.
Hy3 pricing
Official pricing
- $0.035 cached input
- 262K context
- text input
- 5 routes on OpenRouter
- What Hoodgen routes to first
- Same model, lower bill
- Priority or premium tier
- Best under load
- Use for latency critical calls
Market pricing, every route
| Provider | Tier | Input | Output | Cache read | Uptime 24h |
|---|---|---|---|---|---|
| Tencent | standard | $0.083 | $0.33 | $0.021 | 100.0% |
| Novita | standard | $0.14 | $0.58 | $0.035 | 100.0% |
| GMICloud | standard | $0.14 | $0.58 | $0.035 | 100.0% |
| Phala | standard | $0.15 | $0.64 | $0.04 | 100.0% |
| AtlasCloud | standard | $0.2 | $0.8 | $0.05 | 100.0% |
API pricing and cost calculator
| Rate | Per call | Total | Credits |
|---|---|---|---|
| Cheapest route | $0.0014 | $0.70 | 667 |
| List price | $0.0023 | $1.13 | 1078 |
| Highest route | $0.0052 | $2.60 | 2470 |
API and developer information
| Model ID | tencent/hy3 |
| Endpoint | https://openrouter.ai/api/v1/chat/completions |
| Parameters | prompt and max_tokens |
| Streaming | Yes, server sent events |
| Modality string | text->text |
| Max prompt tokens | 262,144 |
| Max completion tokens | 235,929 |
curl https://openrouter.ai/api/v1/chat/completions \
-H "Authorization: Bearer $OPENROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "tencent/hy3",
"messages": [{ "role": "user", "content": "Summarise this document." }],
"max_tokens": 1024
}'Model limitations
| Text only output | It cannot generate images, audio or video. Pair it with an image model on Hoodgen when you need visuals. |
| Output cap | 235,929 tokens per request. Long deliverables have to be chunked. |
| Benchmarks not published | Artificial Analysis does not publish an Intelligence Index for this release, so no benchmark score is claimed here. |
| Knowledge cutoff | The provider does not state a fixed cutoff for every release. Verify very recent events against search. |
Hy3 vs other Tencent models
| Model | Context | Input /1M | Output /1M | AA index | Released |
|---|---|---|---|---|---|
| Hy3 | 262K | $0.132 | $0.528 | n/a | 2026 |
| Hy-MT2-30B-A3B | 8K | $0.074 | $0.295 | n/a | 2026 |
| Hy-MT2-1.8B | 8K | $0.044 | $0.177 | n/a | 2026 |
Privacy and data policy
- On Hoodgen, prompts and attachments answer that request and stay in your conversation history so you can reopen it. They are not used to train models.
- Requests route through OpenRouter to Tencent and are covered by that provider's API terms.
- Do not paste secrets or personal data you would not send to any hosted model.
Hy3 questions, answered
Start chatting with Hy3
Free to try on Hoodgen AI, with a live credit estimate before every send. No API key, no setup.