deepseek-ai/DeepSeek-V4-Pro
Text Generation • 862B • Updated • 174k • • 3.16k
Leco Li PRO
imnotkitty
AI & ML interests
None yet
Recent Activity
liked a model 33 minutes ago
deepseek-ai/DeepSeek-V4-Pro reacted to wangbuer999's post with 🔥 5 days ago
Testing AI controlling AI with Hy3 Preview I barely lifted a finger the whole time.
One-click deployment of Hermes on WorkBuddy took some time with a few rounds of adjustments, and I finally got it up and running smoothly
Only minor issue was setting up Supermemory it was a bit slow on the uptake. I had to go over simple steps several times, guiding it patiently like teaching a kid.
The experience of AI orchestrating AI is absolutely incredible. started running Agents with Hunyuan right after its release, and it actually works perfectly.
295B parameters, 21B active parameters, with direct access to TokenHub now great cost-performance ratio too
Honestly, I used to get stuck on all kinds of environment configurations when deploying Agents locally. Using Hy3 to take command made the whole process way more streamlined. repliedto their post 6 days ago
https://huggingface.co/spaces/tencent/Hy3-preview is out: an open-weights MoE reasoning model.
✅ 295B total / 21B active / 256K context
✅ Fused fast-and-slow thinking in a single model
✅ First model trained on Hunyuan's rebuilt pretraining + RL infra (Feb → Apr)
Benchmarks:
👉 SWE-Bench Verified, Terminal-Bench 2.0, BrowseComp, WideSearch — competitive results, particularly strong on agentic tool use
👉 Top score on Tsinghua's 2026 Spring math PhD qualifying exam
👉 Strong context-learning and instruction-following on Tencent's CL-bench / CL-bench-Life
More details can be found in my article: https://huggingface.co/blog/imnotkitty/hy3-previewOrganizations
liked a
model 33 minutes ago
reacted to wangbuer999's post with 🔥 5 days ago
Post
3534
Testing AI controlling AI with Hy3 Preview I barely lifted a finger the whole time.
One-click deployment of Hermes on WorkBuddy took some time with a few rounds of adjustments, and I finally got it up and running smoothly
Only minor issue was setting up Supermemory it was a bit slow on the uptake. I had to go over simple steps several times, guiding it patiently like teaching a kid.
The experience of AI orchestrating AI is absolutely incredible. started running Agents with Hunyuan right after its release, and it actually works perfectly.
295B parameters, 21B active parameters, with direct access to TokenHub now great cost-performance ratio too
Honestly, I used to get stuck on all kinds of environment configurations when deploying Agents locally. Using Hy3 to take command made the whole process way more streamlined.
One-click deployment of Hermes on WorkBuddy took some time with a few rounds of adjustments, and I finally got it up and running smoothly
Only minor issue was setting up Supermemory it was a bit slow on the uptake. I had to go over simple steps several times, guiding it patiently like teaching a kid.
The experience of AI orchestrating AI is absolutely incredible. started running Agents with Hunyuan right after its release, and it actually works perfectly.
295B parameters, 21B active parameters, with direct access to TokenHub now great cost-performance ratio too
Honestly, I used to get stuck on all kinds of environment configurations when deploying Agents locally. Using Hy3 to take command made the whole process way more streamlined.
replied to their post 6 days ago
My guess is they're keeping a low profile, might be up to something big? 🧐
reacted to their post with 🔥 6 days ago
Post
3921
tencent/Hy3-preview is out: an open-weights MoE reasoning model.
✅ 295B total / 21B active / 256K context
✅ Fused fast-and-slow thinking in a single model
✅ First model trained on Hunyuan's rebuilt pretraining + RL infra (Feb → Apr)
Benchmarks:
👉 SWE-Bench Verified, Terminal-Bench 2.0, BrowseComp, WideSearch — competitive results, particularly strong on agentic tool use
👉 Top score on Tsinghua's 2026 Spring math PhD qualifying exam
👉 Strong context-learning and instruction-following on Tencent's CL-bench / CL-bench-Life
More details can be found in my article: https://huggingface.co/blog/imnotkitty/hy3-preview
✅ 295B total / 21B active / 256K context
✅ Fused fast-and-slow thinking in a single model
✅ First model trained on Hunyuan's rebuilt pretraining + RL infra (Feb → Apr)
Benchmarks:
👉 SWE-Bench Verified, Terminal-Bench 2.0, BrowseComp, WideSearch — competitive results, particularly strong on agentic tool use
👉 Top score on Tsinghua's 2026 Spring math PhD qualifying exam
👉 Strong context-learning and instruction-following on Tencent's CL-bench / CL-bench-Life
More details can be found in my article: https://huggingface.co/blog/imnotkitty/hy3-preview
- 2 replies
upvoted an article 6 days ago
Article
Hy3 preview: A Rebuilt Hunyuan, a 21B-Active MoE, and a New Reasoning Receipe
•
5
published an article 6 days ago
Article
Hy3 preview: A Rebuilt Hunyuan, a 21B-Active MoE, and a New Reasoning Receipe
•
5
posted an update 6 days ago
Post
3921
tencent/Hy3-preview is out: an open-weights MoE reasoning model.
✅ 295B total / 21B active / 256K context
✅ Fused fast-and-slow thinking in a single model
✅ First model trained on Hunyuan's rebuilt pretraining + RL infra (Feb → Apr)
Benchmarks:
👉 SWE-Bench Verified, Terminal-Bench 2.0, BrowseComp, WideSearch — competitive results, particularly strong on agentic tool use
👉 Top score on Tsinghua's 2026 Spring math PhD qualifying exam
👉 Strong context-learning and instruction-following on Tencent's CL-bench / CL-bench-Life
More details can be found in my article: https://huggingface.co/blog/imnotkitty/hy3-preview
✅ 295B total / 21B active / 256K context
✅ Fused fast-and-slow thinking in a single model
✅ First model trained on Hunyuan's rebuilt pretraining + RL infra (Feb → Apr)
Benchmarks:
👉 SWE-Bench Verified, Terminal-Bench 2.0, BrowseComp, WideSearch — competitive results, particularly strong on agentic tool use
👉 Top score on Tsinghua's 2026 Spring math PhD qualifying exam
👉 Strong context-learning and instruction-following on Tencent's CL-bench / CL-bench-Life
More details can be found in my article: https://huggingface.co/blog/imnotkitty/hy3-preview
- 2 replies
liked a
Space 6 days ago
posted an update 8 days ago
Post
170
moonshotai/Kimi-K2.6 just dropped!
A few numbers that stood out:
✅ 13h non-stop coding, 4,000+ lines
✅ 300 sub-agents, 4,000 steps
✅ 5-day autonomous runs (OpenClaw / Hermes)
✅ Parity with GPT-5.4 / Opus 4.6 / Gemini 3.1 Pro on SWE-Bench Pro + HLE
The 5-day autonomous number is the one I'd most like to see reproduced.
A few numbers that stood out:
✅ 13h non-stop coding, 4,000+ lines
✅ 300 sub-agents, 4,000 steps
✅ 5-day autonomous runs (OpenClaw / Hermes)
✅ Parity with GPT-5.4 / Opus 4.6 / Gemini 3.1 Pro on SWE-Bench Pro + HLE
The 5-day autonomous number is the one I'd most like to see reproduced.
liked a
model 8 days ago
reacted to prithivMLmods's post with 👀 9 days ago
Post
4121
HY-World-2.0 — A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds is now available on Spaces, and it works both as native Gradio components and in Gradio server mode.
> HY-World-2.0-Demo: prithivMLmods/HY-World-2.0-Demo
> HY-World-2.0 [Server Mode]: prithivMLmods/HY-World-2.0-Demo
> Featuring 3D reconstruction and Gaussian splats with the Rerun viewer, along with camera poses, depth maps, and surface normals.
> In Server Mode, Gradio is served via FastAPI, with FastAPI remaining the top-level server.
> Model: tencent/HY-World-2.0
> GitHub: https://github.com/PRITHIVSAKTHIUR/HY-World-2.0-Demo
🤗To learn more, visit the app page or the respective model pages.
> HY-World-2.0-Demo: prithivMLmods/HY-World-2.0-Demo
> HY-World-2.0 [Server Mode]: prithivMLmods/HY-World-2.0-Demo
> Featuring 3D reconstruction and Gaussian splats with the Rerun viewer, along with camera poses, depth maps, and surface normals.
> In Server Mode, Gradio is served via FastAPI, with FastAPI remaining the top-level server.
> Model: tencent/HY-World-2.0
> GitHub: https://github.com/PRITHIVSAKTHIUR/HY-World-2.0-Demo
🤗To learn more, visit the app page or the respective model pages.
reacted to ajibawa-2023's post with 👍 9 days ago
Post
6073
Go-Code-Large
Dataset: ajibawa-2023/Go-Code-Large
Go-Code-Large is a large-scale corpus of Go (Golang) programming language source code, comprising 316,427 code samples stored in .jsonl format. The dataset is designed to support research and development in large language model (LLM) pretraining, static analysis, cloud-native systems, and modern backend software engineering.
By offering a focused and curated dataset for Go, this corpus enables experimentation in concurrent programming, distributed systems, and performance-oriented backend services—domains where Go is widely adopted.
Go-Code-Large addresses the relative scarcity of large, language-specific datasets for Go, enabling targeted research into idiomatic Go patterns, concurrency primitives, and scalable system design.
Dataset: ajibawa-2023/Go-Code-Large
Go-Code-Large is a large-scale corpus of Go (Golang) programming language source code, comprising 316,427 code samples stored in .jsonl format. The dataset is designed to support research and development in large language model (LLM) pretraining, static analysis, cloud-native systems, and modern backend software engineering.
By offering a focused and curated dataset for Go, this corpus enables experimentation in concurrent programming, distributed systems, and performance-oriented backend services—domains where Go is widely adopted.
Go-Code-Large addresses the relative scarcity of large, language-specific datasets for Go, enabling targeted research into idiomatic Go patterns, concurrency primitives, and scalable system design.
- 2 replies
reacted to their post with 🔥 13 days ago
Post
2595
Just tried tencent/HY-World-2.0 — a multimodal world model that takes in text or a single image and generates editable 3D scenes.
Unlike Google's Genie and HY-World 1.5, v2.0 generates engine-ready 3D content:
🎮 Direct import into Unreal Engine and Unity — no format wrangling
🧊 Supports multiple 3D asset formats: Mesh, 3DGS, point cloud, etc.
✏️ Fully editable — not a baked video, but actual geometry you can modify
🤖 Also usable for embodied simulation environments
Basically: from "AI generates a world you can look at" → "AI generates a world you can ship."
Unlike Google's Genie and HY-World 1.5, v2.0 generates engine-ready 3D content:
🎮 Direct import into Unreal Engine and Unity — no format wrangling
🧊 Supports multiple 3D asset formats: Mesh, 3DGS, point cloud, etc.
✏️ Fully editable — not a baked video, but actual geometry you can modify
🤖 Also usable for embodied simulation environments
Basically: from "AI generates a world you can look at" → "AI generates a world you can ship."
posted an update 13 days ago
Post
2595
Just tried tencent/HY-World-2.0 — a multimodal world model that takes in text or a single image and generates editable 3D scenes.
Unlike Google's Genie and HY-World 1.5, v2.0 generates engine-ready 3D content:
🎮 Direct import into Unreal Engine and Unity — no format wrangling
🧊 Supports multiple 3D asset formats: Mesh, 3DGS, point cloud, etc.
✏️ Fully editable — not a baked video, but actual geometry you can modify
🤖 Also usable for embodied simulation environments
Basically: from "AI generates a world you can look at" → "AI generates a world you can ship."
Unlike Google's Genie and HY-World 1.5, v2.0 generates engine-ready 3D content:
🎮 Direct import into Unreal Engine and Unity — no format wrangling
🧊 Supports multiple 3D asset formats: Mesh, 3DGS, point cloud, etc.
✏️ Fully editable — not a baked video, but actual geometry you can modify
🤖 Also usable for embodied simulation environments
Basically: from "AI generates a world you can look at" → "AI generates a world you can ship."
liked a
model 13 days ago
posted an update 27 days ago
Post
132
Qwen3.6-Plus is now available — near-Claude-level performance
Key highlights:
✅ 1M context
✅ One-shot Vibe Coding (3D scenes, games, full-stack apps)
✅ Outperforms models 2-3x its parameter size
✅ Strong Vibe Coding results: 3D scenes, interactive games, full websites from single prompts
✅ Optimized for Agent workflows (OpenClaw, Qwen Code, Cline, etc.)
API cost is significantly lower than comparable models.
Key highlights:
✅ 1M context
✅ One-shot Vibe Coding (3D scenes, games, full-stack apps)
✅ Outperforms models 2-3x its parameter size
✅ Strong Vibe Coding results: 3D scenes, interactive games, full websites from single prompts
✅ Optimized for Agent workflows (OpenClaw, Qwen Code, Cline, etc.)
API cost is significantly lower than comparable models.
reacted to SeaWolf-AI's post with 🔥 about 1 month ago
Post
4684
🌍 World Model Bench — does your world model actually think?
FID measures realism. FVD measures smoothness. But neither tells you whether the model understood the scene.
We just released WM Bench — the first benchmark for cognitive intelligence in world models. The core question: when a beast charges from 3 meters away, does the model know to sprint — not walk? Does it respond differently to a human vs an animal? Does it remember the left corridor was blocked two steps ago?
Those are cognitive questions. No existing benchmark asks them. So we built one.
3 Pillars · 10 Categories · 100 Scenarios · 1,000-point scale
- 👁 P1 Perception (25%) — Can it read the scene?
- 🧠 P2 Cognition (45%) — Does it predict threats, escalate emotions, utilize memory?
- 🔥 P3 Embodiment (30%) — Does the body respond with the right motion?
All evaluation is via simple JSON I/O — no 3D engine, no special hardware. Any model with an API can participate.
We also built PROMETHEUS as a live reference implementation — runs in your browser on a T4, no install needed. Combines FloodDiffusion motion generation with a LLM cognitive brain (Perceive → Predict → Decide → Act). Scored 726/1000 (Grade B) on Track C — the only directly verified model so far. Submissions from other teams very welcome.
---
🗂 Dataset → FINAL-Bench/World-Model
🌍 Demo → FINAL-Bench/World-Model
🏆 Leaderboard → FINAL-Bench/worldmodel-bench
📝 Article → https://huggingface.co/blog/FINAL-Bench/world-model
Part of the FINAL Bench Family — alongside FINAL Bench (Feb 2026). Feedback on rubrics and missing models always welcome!
FID measures realism. FVD measures smoothness. But neither tells you whether the model understood the scene.
We just released WM Bench — the first benchmark for cognitive intelligence in world models. The core question: when a beast charges from 3 meters away, does the model know to sprint — not walk? Does it respond differently to a human vs an animal? Does it remember the left corridor was blocked two steps ago?
Those are cognitive questions. No existing benchmark asks them. So we built one.
3 Pillars · 10 Categories · 100 Scenarios · 1,000-point scale
- 👁 P1 Perception (25%) — Can it read the scene?
- 🧠 P2 Cognition (45%) — Does it predict threats, escalate emotions, utilize memory?
- 🔥 P3 Embodiment (30%) — Does the body respond with the right motion?
All evaluation is via simple JSON I/O — no 3D engine, no special hardware. Any model with an API can participate.
We also built PROMETHEUS as a live reference implementation — runs in your browser on a T4, no install needed. Combines FloodDiffusion motion generation with a LLM cognitive brain (Perceive → Predict → Decide → Act). Scored 726/1000 (Grade B) on Track C — the only directly verified model so far. Submissions from other teams very welcome.
---
🗂 Dataset → FINAL-Bench/World-Model
🌍 Demo → FINAL-Bench/World-Model
🏆 Leaderboard → FINAL-Bench/worldmodel-bench
📝 Article → https://huggingface.co/blog/FINAL-Bench/world-model
Part of the FINAL Bench Family — alongside FINAL Bench (Feb 2026). Feedback on rubrics and missing models always welcome!
liked a
model about 1 month ago
posted an update about 1 month ago
Post
196
🚀 GLM-5-Turbo just dropped: The "OpenClaw Native" Model
Z.ai just shipped a 744B MoE beast that's 2-3x faster than GLM-5, with 200K context + 128K max output.
What's different:
👉 Tool call stability (no more random failures mid-chain)
👉 Complex instruction decomposition (breaks down messy prompts)
👉 Time-aware execution (understands scheduled/persistent tasks)
👉 High-throughput long-chain efficiency (doesn't choke on 50-step workflows)
📊 ZClawBench: Leads mainstream models in OpenClaw scenarios
💰 Trade-off: +20% price vs GLM-5
Anyone trying it yet? 👀
Z.ai just shipped a 744B MoE beast that's 2-3x faster than GLM-5, with 200K context + 128K max output.
What's different:
👉 Tool call stability (no more random failures mid-chain)
👉 Complex instruction decomposition (breaks down messy prompts)
👉 Time-aware execution (understands scheduled/persistent tasks)
👉 High-throughput long-chain efficiency (doesn't choke on 50-step workflows)
📊 ZClawBench: Leads mainstream models in OpenClaw scenarios
💰 Trade-off: +20% price vs GLM-5
Anyone trying it yet? 👀
- 1 reply
reacted to branikita's post with 🚀 about 1 month ago
Post
1871
Robonine just published a new article! Mechanical backlash is a common limitation in servo-driven robotic joints. In this experiment, paired Feetech STS3215 servos are used with a small opposing preload to eliminate gearbox play, significantly improving positional stability and motion precision in robotic manipulators.
https://robonine.com/backlash-compensation-in-sts3215-servo-actuators/
https://robonine.com/backlash-compensation-in-sts3215-servo-actuators/
- 2 replies