Hy3 preview
Tencent · released Apr 13, 2026 · tencent/Hy3-preview
- Type
- Open weightsCustom licence
- Params
- 299B
- Context
- 262K
active per word not recorded by us · about 197K words of context · download allowed, licence restricts use
Our take
Written Sep 1, 2026Hy3 preview is a 299-billion-parameter text-only model from Tencent with a 262,144-token request limit and a custom restricted licence. It scores highest on coding prompts but weakest on creative writing, making it a specialist pick rather than an all-rounder.
Choose this for coding workloads where its measured coding score is the headline strength, or for long-context text tasks needing a quarter-million-token window. It suits budget inference through OpenRouter or GMI Cloud. Skip it if you need creative writing, a permissive licence, or strong web-development build tasks.
The case for it
- Strong measured performance on coding prompts — 105.99 points above its creative writing score, its widest subscore gap.
- Competitive on hard prompts and mathematics, both exceeding its overall text score.
- 262,144-token request limit with no active-parameter reduction verified.
The case against it
- Creative writing is a clear weak point, its lowest subscore and 58.13 points below even web-development coding.
- Web development coding trails general coding by 104.6 points, suggesting build-task weakness.
- Custom licence with open-restricted weights — terms require legal review before fine-tuning or redistribution.
How good is it?
EverydayGeneral questions and everyday reasoning
Arena Text (overall)91st of 168 · 1409
CodingWriting and fixing code on its own
Arena Coding91st of 168 · 1453
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Arena Creative Writing99th of 168 · 1354
Arena Creative Writing is the only board that has scored it for this.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done.
Every published score for this model7 scoresEvery figure we hold, from 7 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Comfortable fit
Apple M3 Ultra (80-core GPU) · 512 GB
Room to spare. 188.1 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked 4 hours ago — each listing carries its own date.
- per 1M tokens
- $0.18 in / $0.60 out
- Context served
- 262K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $0.18 / $0.60checked 4 hours ago | 262K | not measured | Unknown | Unknown | Unknown |
| GMICloudbf16Through OpenRouter | $0.18 / $0.60checked 4 hours ago | 262K236K max reply | 83 tok/s | No | Yesunknown period | Unknown |
Across the 2 listings we hold: 1 says it does not train on prompts, 0 say they do and 1 does not say. 0 appear in the zero-retention registry we check; the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouterOpenRouter's own listing | ✓ | ✗ | ✗ |
| GMICloudbf16Through OpenRouter | ✓ | ✗ | ✗ |
Tool calling: 2 of 2 listings say yes. JSON output: 0 of 2 listings say yes, 2 say no. Strict schema: 0 of 2 listings say yes, 2 say no.
Models people weigh against Hy3 preview
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- We do not hold the active parameter count for it, so how much of it runs on any one token is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 2 listings does not say whether it trains on prompts.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsCustom licence, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Custom licence
This model ships custom license terms that don't map to a known template. We haven't parsed them, so commercial use, redistribution and derivatives are unverified — review the original terms before shipping.
Identifiers
- Hugging Face
- tencent/Hy3-preview
- Architecture
- Mixture of experts
- Takes in, gives back
- Text in, text out
- Catalogue slug
- tencent-hy3-preview