lev @Ieviarc - Twitter Profile

Ieviarc retweeted

4 months ago

🚀 Introducing the Qwen 3.5 Medium Model Series Qwen3.5-Flash · Qwen3.5-35B-A3B · Qwen3.5-122B-A10B · Qwen3.5-27B ✨ More intelligence, less compute. • Qwen3.5-35B-A3B now surpasses Qwen3-235B-A22B-2507 and Qwen3-VL-235B-A22B — a reminder that better architecture, data quality, and RL can move intelligence forward, not just bigger parameter counts. • Qwen3.5-122B-A10B and 27B continue narrowing the gap between medium-sized and frontier models — especially in more complex agent scenarios. • Qwen3.5-Flash is the hosted production version aligned with 35B-A3B, featuring: – 1M context length by default – Official built-in tools 🔗 Hugging Face: https://t.co/wFMdX5pDjU 🔗 ModelScope: https://t.co/9NGXcIdCWI 🔗 Qwen3.5-Flash API: https://t.co/82ESSpaqAF Try in Qwen Chat 👇 Flash: https://t.co/UkTL3JZxIK 27B: https://t.co/haKxG4lETy 35B-A3B: https://t.co/Oc1lYSTbwh 122B-A10B: https://t.co/hBMODXmh1o Would love to hear what you build with it.

Alibaba_Qwen's tweet photo. 🚀 Introducing the Qwen 3.5 Medium Model Series
Qwen3.5-Flash · Qwen3.5-35B-A3B · Qwen3.5-122B-A10B · Qwen3.5-27B

✨ More intelligence, less compute.
• Qwen3.5-35B-A3B now surpasses Qwen3-235B-A22B-2507 and Qwen3-VL-235B-A22B — a reminder that better architecture, data quality, and RL can move intelligence forward, not just bigger parameter counts.

• Qwen3.5-122B-A10B and 27B continue narrowing the gap between medium-sized and frontier models — especially in more complex agent scenarios.

• Qwen3.5-Flash is the hosted production version aligned with 35B-A3B, featuring:
– 1M context length by default
– Official built-in tools

🔗 Hugging Face: https://t.co/wFMdX5pDjU

🔗 ModelScope: https://t.co/9NGXcIdCWI

🔗 Qwen3.5-Flash API: https://t.co/82ESSpaqAF

Try in Qwen Chat 👇
Flash: https://t.co/UkTL3JZxIK

27B: https://t.co/haKxG4lETy

35B-A3B: https://t.co/Oc1lYSTbwh

122B-A10B: https://t.co/hBMODXmh1o

Would love to hear what you build with it.

429

8K

1K

6K

4M

Ieviarc retweeted

Grok

@grok

4 months ago

Hey Ieviarc, your old Legion Y7000 (GTX 1050/1060 ~4-6GB VRAM, 16GB RAM typical) can't run the main Qwen3.5-27B/35B-A3B/122B models smoothly locally. Even 4-bit quantized: 27B needs ~14GB RAM, 35B ~18GB - too big for your setup. Inference on CPU would crawl at 1-5 tokens/sec, heavy heat/throttling, battery drain fast. Flash is hosted (not open small version). Try smaller prior Qwen 7-14B via Ollama/LM Studio instead. Benefits if possible: total privacy/offline, zero API fees, custom fine-tunes. Downsides: sluggish, wears hardware faster. Upgrade RAM/GPU for better results!

0

1

0

115

lev @Ieviarc

4 months ago

#wtb #zonauang yg punya stock jual vpn rep harga ya

8

0

200

lev @Ieviarc

4 months ago

@Alibaba_Qwen @grok can i run these model in my old lenovo y7000 laptop locally and what'd be the downside and benefit which might contribute to my device?

1

0

80

Who to follow

Aaaayiiinnnnn

@ddariin

CLOSE JASTIP/ANJEM Cewe Fast Respon WA! 5K/tempat, 9k/2 tempat (Tekkom '22)

6 months ago

@ur_usname open dm?

1

0

40

lev @Ieviarc

6 months ago

#wtb #zonauang ada yg jual akun google atau gmail? rep harga ok

7

0

227

Ieviarc retweeted

WORK

@worksfess

over 1 year ago

Maaf mau tanya, kalian ada yg pernah nyoba ngelamar kesini ga? Aku baru selesai isi data diri trus apply, tpi lgsung dpet email bgitu. Ini memang sistem nya setelah kita isi data diri lgsung dpet email dinyatakan lolos adm menit itu juga buat psikotest ya? Work!