Tencent @TencentHunyuan dropped the non-preview version of Hy3 and changed their license from the community one (restrictive + not allowed in SK, UK, EU) to Apache 2.0!!! 🙏
10:58 PM · Jul 5, 2026 · 39.7K Views
Tencent releases Hunyuan3 under Apache 2.0 license, lifting geographic restrictions on its 295B parameter MoE model
Early testing shows the model matches GLM-5.1 on agentic benchmarks.
8699270244179K
Sentiment
Many users praised Tencent's Hy3 release under Apache 2.0 for unlocking enterprise use and open-model momentum while a few called the model inferior to alternatives like GLM.
Pos
90.5%
Neg
9.5%
19 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Related links
Posts from X
Most Activity
Most Activity
VIEWS57.4KBOOKMARKS90LIKES296REPLIES23
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
If these scores are real, Tencent has just become one of the leaders of open source. Hy3 is a 295B 21AB model that at least beats GLM 5.1 in blind tests. Might even warrant architecture transition (it's a basic GQA). Good job.
Florian Brand@xeophon
Tencent @TencentHunyuan dropped the non-preview version of Hy3 and changed their license from the community one (restrictive + not allowed in SK, UK, EU) to Apache 2.0!!! 🙏
4hViews 57.4KLikes 296Bookmarks 90
RETWEETS21
elie@eliebakouch
new open (apache 2.0!) model from @TencentHunyuan, Hy3 is only 295B total 21B active and competitive with MUCH bigger model on benchmarks
https://huggingface.co/tencent/Hy3
4hViews 27.8KLikes 243Bookmarks 68
elie@eliebakouch
the benchmark scores are REALLY crazy for a 300B model
elie@eliebakouch
new open (apache 2.0!) model from @TencentHunyuan, Hy3 is only 295B total 21B active and competitive with MUCH bigger model on benchmarks
https://huggingface.co/tencent/Hy3
3hViews 15.9KLikes 153Bookmarks 42
xjdr@_xjdr
Ohh this looks incredibly promising. Time to throw some internal evals at it . If it's performance is even moderately close to it's published evals, this could be a really great model
Florian Brand@xeophon
Tencent @TencentHunyuan dropped the non-preview version of Hy3 and changed their license from the community one (restrictive + not allowed in SK, UK, EU) to Apache 2.0!!! 🙏
3hViews 6.9KLikes 58Bookmarks 9
Zephyr@zephyr_z9
Pretty strong model from Tencent
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
If these scores are real, Tencent has just become one of the leaders of open source. Hy3 is a 295B 21AB model that at least beats GLM 5.1 in blind tests. Might even warrant architecture transition (it's a basic GQA). Good job.
3hViews 20.1KLikes 45Bookmarks 8
xjdr@_xjdr
Please hold, this might be replacing dsv4-flash for me . More testing needed but early results are surprisingly promising
xjdr@_xjdr
Ohh this looks incredibly promising. Time to throw some internal evals at it . If it's performance is even moderately close to it's published evals, this could be a really great model
1hViews 3KLikes 27Bookmarks 5
elie@eliebakouch
from the model card seems to not only be benchmark
> We don't think public benchmark scores tell the full story. So we ran a blind test with 270 experts from various disciplines, working on real-world workflows, and collected 312 valid comparisons. Hy3 scored 2.67/4, outperforming GLM-5.1 at 2.51/4. The advantage was clearest in frontend development, CI/CD, and data & storage.
elie@eliebakouch
new open (apache 2.0!) model from @TencentHunyuan, Hy3 is only 295B total 21B active and competitive with MUCH bigger model on benchmarks
https://huggingface.co/tencent/Hy3
4hViews 4.4KLikes 25Bookmarks 0
Florian Brand@xeophon
@Jaidcel @TencentHunyuan Yes! https://huggingface.co/tencent/Hy3
3hViews 441Likes 4Bookmarks 3
elie@eliebakouch
@teortaxesTex doesn't have 1M context! model without sparse attention is ok at 256k (also you can do GQA -> MSA)
3hViews 125Likes 5Bookmarks 2
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
@eliebakouch There are ways to convert GQA to MLA and therefore to DSA I mean that this model is strong but with the current baselines not economical for long sequences
3hViews 551Likes 3Bookmarks 2
Florian Brand@xeophon
@teortaxesTex Some truly insane jumps, especially for this size
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
If these scores are real, Tencent has just become one of the leaders of open source. Hy3 is a 295B 21AB model that at least beats GLM 5.1 in blind tests. Might even warrant architecture transition (it's a basic GQA). Good job.
3hViews 7.8KLikes 13Bookmarks 0
Jaid@Jaidcel
@xeophon @TencentHunyuan Updated weights?
3hViews 446Likes 3Bookmarks 1
elie@eliebakouch
@teortaxesTex wdym? > Might even warrant architecture transition (it's a basic GQA)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
If these scores are real, Tencent has just become one of the leaders of open source. Hy3 is a 295B 21AB model that at least beats GLM 5.1 in blind tests. Might even warrant architecture transition (it's a basic GQA). Good job.
3hViews 1.5KLikes 5Bookmarks 0
xlr8harder@xlr8harder
We love to see it!
Florian Brand@xeophon
Tencent @TencentHunyuan dropped the non-preview version of Hy3 and changed their license from the community one (restrictive + not allowed in SK, UK, EU) to Apache 2.0!!! 🙏
2hViews 704Likes 8Bookmarks 0
Zephyr@zephyr_z9
@_xjdr btw, new DSv4 is launching soon (likely this week)
58mViews 527Likes 8
🜑@1casie
@xeophon @TencentHunyuan >not allowed in SK, UK, EU
3hViews 269Likes 3
dinos@din0s_
@xeophon you consistently naming and shaming non apache releases as soon as they come out has probably had a non negligible impact on the state of open models, good job
2hViews 198Likes 3
Florian Brand@xeophon
@din0s_ haha, idk how much of an impact it really has, i just don't want to deal with reading licenses :(
2hViews 169Likes 2
naeem@identity_matrix
@eliebakouch @TencentHunyuan Wow, it has improved a lot from preview.
Makes me even more excited about deepseek v4 non-preview release
3hViews 235Likes 5
elie@eliebakouch
@_xjdr Please share the result 🫡
xjdr@_xjdr
Ohh this looks incredibly promising. Time to throw some internal evals at it . If it's performance is even moderately close to it's published evals, this could be a really great model
3hViews 392Likes 4Bookmarks 0
Load more posts