tweetindex
ID

ModelScope ✓

@ModelScope2022 · HangZhou, China · joined 28 Apr 2024

Driving innovations with open communities. 💬 Join our Discord: https://t.co/4M9AHVh5qa

14 409Followers
168Following
1 056Posts total
152.7KViews on collected posts

Postingan terbaru

OpenVDN releases VDN-H3, a faster-than-playback adaptation of MiniMax-H3. 🤖 https://t.co/fcAL0EqzGj ⚡ The 8-step model reaches 11.23s steady-state denoising on the reported 8-GPU setup, excluding loading, warm-up, VAE decoding, and video encoding. 🧠 Its hybrid architecture http 1.9K views · 25 likes · 3 reposts · 2 replies Open on X →
Just 0.199% trainable parameters turn a 33B video generator into a controllable world. 🎮 Meet H3-World. 🤖 https://t.co/013ZCwN9E4 📄 https://t.co/SQCXJABGka 🕹️ Starting from one frame, keyboard sequences steer both character and camera motion to produce different futures. ⏱️ http 1.2K views · 7 likes · 0 reposts · 2 replies Open on X →
Ling-3.0-flash-Fin and FinFIRST are now open: a finance model built for real workflows, plus a benchmark that verifies the work. Developed by Ant Group with financial institutions and domain experts.🏦 🧠 MODEL 🤖 https://t.co/9eRBs7ab7f ⚙️ 124B MoE, only 5.1B active per token, http 4.5K views · 71 likes · 4 reposts · 1 replies Open on X →
Say hello to DreamX-Creator 1.0, a 7B native audio-video generator with local 2K refinement. Apache 2.0. 🤖 https://t.co/3bFWEDRgfy 🏆 Delivers results comparable to much larger SOTA models like MiniMax H3 on Verse Bench, reaching 0.1351 DeSync after RL and 0.6930 VQ with the 2K h 22.9K views · 276 likes · 35 reposts · 12 replies Open on X →
@ModelScope2022 4b model outperforms 9b models .wow 😲🤯 842 views · 5 likes · 0 reposts · 0 replies Open on X →
@ModelScope2022 @grok analisa os specs do modelo e verifica o hardware mínimo para o rodar. 1.5K views · 4 likes · 0 reposts · 1 replies Open on X →
@ModelScope2022 4B beating a 9B model on swe-bench pro is wild, apache 2.0 too so no waitlist nonsense 602 views · 4 likes · 0 reposts · 0 replies Open on X →
Agent capabilities in just 4B parameters, with native 1M context. Don’t miss Spark-X2.5-4B. Apache 2.0. 🤖 https://t.co/BT83fxmtLz 🏆 Beats Qwen3.5-9B on SWE-bench Pro (44.4 vs. 33.8) and SWE-bench Multilingual (53.3 vs. 43.3). 🛠️ Leads the comparison on τ³-bench, MCP-Atlas, htt 49.1K views · 588 likes · 48 reposts · 31 replies Open on X →
Breeze-TTS-2 is here. 🎙️ Real-time voice cloning, design, and direction in one model. 🤖 https://t.co/TSO2etFJOw 🏆 #1 open-weight TTS on Artificial Analysis, plus first place on voice design, instruction following, and latency benchmarks. 🎭 Clone a voice while preserving https:// 15.2K views · 251 likes · 24 reposts · 7 replies Open on X →
Hy4 preview just got a lot smaller! AngelSlim releases two GGUF builds for llama.cpp, compressing the 770B flagship to 435 GiB or 214 GiB. 🤖 https://t.co/28w6b0jyyG 📦 Q4_K_M at 435.20 GiB is the quality-first default. STQ1_0 cuts that to 213.66 GiB with mixed 1–2 bit 2.5K views · 28 likes · 5 reposts · 3 replies Open on X →
🌍 For global users: https://t.co/VDYKgPeOP4 921 views · 3 likes · 0 reposts · 0 replies Open on X →
@ModelScope2022 deepseek only company with so many variations for the same v4 generation. everybody else going v1,v2,v3,v4.. i think when deepseek release new v5 version, that will be Fable level for sure. 360 views · 4 likes · 0 reposts · 0 replies Open on X →
DeepSeek-V4 can see now. 👀 DeepSeek-V4-Flash-Vision-Exp is the family’s first experimental multimodal model. MIT License. 🤖 https://t.co/U8mYcNGHqV 🏆 Beats Opus-4.8 on Agents’ Last Exam and ZeroBench, and trails it by just 0.7 on Chartography. 🖼️ Understands images, https://t.co 49K views · 241 likes · 20 reposts · 10 replies Open on X →
A new robotics paper worth a look: WALL-SS explores long-horizon world models for robot simulation and policy evaluation. 📄 https://t.co/tIWU7KBlPV What stands out: 🏆 #1 across four simulation metrics on 300 seen and unseen tasks. 📊 Simulated outcomes closely track reality, http 2.3K views · 33 likes · 2 reposts · 6 replies Open on X →

Dibandingkan akun berukuran sama

14 postingan dari 90 hari terakhir, dibandingkan dengan rentang 10K–100K pengikut. tampil ke lebih banyak orang dibanding akun seukurannya.

Median tayangan2 114akun ini924median untuk 10K–100K
Jangkauan, %14.67%akun ini3.62%median untuk 10K–100K
Interaksi, %1.25%akun ini1.52%median untuk 10K–100K
MetrikAkun iniMedian untuk 10K–100KRasio
Median tayangan per postingan2 1149242.29×
Jangkauan (tayangan ÷ pengikut)14.67%3.62%4.05×
Tingkat interaksi1.25%1.52%0.82×

Akun lain pada rentang ini →   Bandingkan dengan akun lain →   Bagaimana tolok ukur ini disusun →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

2.3K29 Aug
49K31 Aug
360
9211 Sep
2.5K
15.2K
49.1K2 Sep
602
1.5K
842
22.9K3 Sep
4.5K
1.2K4 Sep
1.9K

Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

1.79%29 Aug
0.57%31 Aug
1.11%
0.33%1 Sep
1.44%
1.89%
1.38%2 Sep
0.66%
0.33%
0.59%
1.45%3 Sep
1.69%
0.76%4 Sep
1.55%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes55.2%1 540 in total
Reposts5.1%141 in total
Replies2.7%75 in total
Quotes1.2%34 in total
Bookmarks35.9%1 001 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Akun serupa

Frequently asked questions about @ModelScope2022

How many followers does @ModelScope2022 have?

14 409 followers.

How many posts has @ModelScope2022 published?

1 056 posts.

How do posts by @ModelScope2022 perform?

Collected posts: 14. Views: 152 697. Likes: 1 540.