Tencent · Chat
Hy3 (Deep Infra)
Tencent's 295B Mixture-of-Experts model with 21B active parameters, native 256K context support, and three reasoning modes for complex reasoning and agentic tasks.
- llminfo IQ
- 127.2
- Market median price
- $0.25 blended / 1M
- Max context
- 262K
- Listings
- 38
Evidence and sources
Advantages and trade-offs
Compared with 10 chat models, picked mainly by popularity and IQ
Advantages
Costs less than most peers
Market median price: $0.25 blended / 1M, cheaper than 9 of 9 priced peers.
Open weights available
The weights are public. 3 of the 7 peers that say either way have open weights.
Trade-offs
Shorter context than most peers
Input & output
as stated by the model's providersCapabilities
Supported operations and provider-stated features
Listings
| 256K | $0.00 | $0.00 | — | — | |
| 262K | $0.066cache read $0.029 | $0.26 | — | — | |
| 262K | $0.13cache read $0.033 | $0.53 | — | — | |
| 262K | $0.13 | $0.53 | — | — | |
| 262K | $0.132cache read $0.033 | $0.528 | — | — | |
| 262K | $0.132cache read $0.033 | $0.528 | — | — | |
| 262K | $0.132cache read $0.033 | $0.528 | 46 | 1,546ms | |
| 262K | $0.132cache read $0.033 | $0.528 | — | — | |
| 262K | $0.132cache read $0.033 | $0.528 | — | — | |
| 262K | $0.14cache read $0.035 | $0.58 | — | — |
Showing 10 of 34: the cheapest per 1M input tokens from each provider.
Also sold by Aihubmix, Aimlapi, Atlascloud, Crossmodel, DeepInfra, Flowdot, Gmicloud, Jalapeno, Kenari, Llmgateway Providers, Novita AI, Opencode Go, Qiniu, Redpill, Requesty, Vercel, Zenmux.
4 listings without a price are hidden.
Included in plans
Benchmarks
Named effort levels run from lowest to highest. Select a result to see its source details.
Human preference
| Benchmark | Score |
|---|---|
| Arena ELO | |
| Chinese | |
| Creative Writing | |
| English |
Show 43 more benchmarksHide 43 benchmarks
Long context
Reasoning
Agentic
Coding
Math
Science
Instruction following
Factuality
Other
Show 3 more resultsHide 3 results
Show 1 more resultHide 1 result
Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · LMArena, CC BY 4.0 (licence) · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated · BullshitBench by Peter Gostev, MIT (licence) · Hugging Face official benchmark leaderboards; licence varies by benchmark · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence)
Around the web
- Discussion
- Discussion
- Hy ASR 3.0: Tencent's LLM speech recognizer for Mandarin, dialects and context, in depth (30 min)Discussion
youtube.com
- Discussion
Also known as: aihubmix/hy3, deepinfra/tencent/hy3, gmi/tencent/hy3, gmi/tencent/hy3-preview, hy3-free, hy3-preview, hy3-preview-free, novita/hy3, novita/tencent/hy3, opencode_go/hy3, tencent/hy3, tencent/hy3-free, tencent/hy3-preview
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works