- Type
- Open weightsApache License 2.0
- Params
- 299B
- Context
- 262K
about 197K words of context
Our take
Written Aug 2, 2026Hy3 is a large downloadable text model from Tencent with a permissive Apache licence and a 262,144-token request limit. It scores notably higher on coding and web-development tasks than on general chat, though creative writing and agentic work are weaker spots.
Choose this for long-context coding, reasoning or instruction-following tasks where you need a permissive licence. It suits budget-conscious coding workloads, and any deployment requiring commercial modification or redistribution rights. Skip it if you need creative writing quality, agentic task performance, or multimodal input.
The case for it
- Coding score is 45.4 points above its general chat score on the Arena leaderboard.
- Web-development coding peaks another 14.7 points above its general coding score.
- Apache 2.0 licence permits commercial use, modification and redistribution.
- 262,144-token request limit handles long documents.
The case against it
- Creative writing score sits 30 points below its own general chat score.
- Agentic task performance is negative on measured runs, the only negative scores in its benchmark set.
- No active parameter count disclosed, so efficiency claims are unverified.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)31st of 143 · 1456.3
CodingWriting and fixing code on its own
Arena Coding35th of 143 · 1500.8
AgenticPlanning, calling tools, staying on task
Arena Agent (IPS)22nd of 36 · −0.01
Arena Agent (IPS) is the only board that has scored it for this.
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Hy3 placed and give it no mark out of five.
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, which is why they get no rating.
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?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
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.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 8 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.13 in / $0.53 out
- Context served
- 262K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $0.13 / $0.53 | 262K | not measured | Unknown | Unknown | Unknown |
| Tencentfp8 | $0.13 / $0.53 | 262K | 49 tok/s | No | No | Confirmed |
| GMICloudbf16 | $0.13 / $0.53 | 262K | 79 tok/s | No | Yesunknown period | Unknown |
| DeepInfrafp8 | $0.14 / $0.58 | 262K | 48 tok/s | No | No | Confirmed |
| Novita AI | $0.14 / $0.58 | 262K | not measured | Unknown | Unknown | Unknown |
| Novita AI | $0.14 / $0.58 | 262K | 50 tok/s | No | No | Confirmed |
| DeepInfrafp8 | $0.14 / $0.58 | 262K | not measured | Unknown | Unknown | Unknown |
| AtlasCloudfp8 | $0.20 / $0.80 | 262K | 120 tok/s | No | Yesunknown period | Unknown |
DeepInfra
Novita AIAcross the 8 listings we hold: 5 say they do not train on prompts, 0 say they do and 3 do not say. 3 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| Tencentfp8 | ✓ | ✓ | ✗ |
| GMICloudbf16 | ✗ | ✓ | ✗ |
| DeepInfrafp8 | ✓ | ✗ | ✓ |
| Novita AI | |||
| Novita AI | ✓ | ✗ | ✓ |
| DeepInfrafp8 | |||
| AtlasCloudfp8 | ✓ | ✗ | ✓ |
Tool calling: 5 of 8 listings say yes, 1 says no, 2 publish no parameter list. JSON output: 3 of 8 listings say yes, 3 say no, 2 publish no parameter list. Strict schema: 4 of 8 listings say yes, 2 say no, 2 publish no parameter list.
Models people weigh against Hy3
When we formed this view
Dates behind this page
Prices last checked 34h ago
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.
- 2 of 8 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 3 of 8 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsApache License 2.0, 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
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- tencent/Hy3
- Architecture
- Mixture of experts
- Modality record
- text->text
- Catalogue slug
- tencent-hy3