@_ryu15_

Local LLMを中心に、AI技術系の投稿とネタツイ Tesla v100 32GB×2枚と共に生きる。 最近はFlash Nextちゃんと遊んでいます アイコンは羊野郎さん@maton_3110

 Qwen3.8 27Bから浮気中
github.com/kuraneko1
Joined August 2019
  • Tweets15.9K
  • Following1.9K
  • Followers7.2K
648Media
Pinned Tweet
Ryu
@_ryu15_Sep 15
完成！！！！！ 僕のVRAM 64GB ヘッドレスAIマシンです！！ マザボ:ASUS Z170-A CPU: intel core i7 6700k GPU: Nvidia v100 32GB + PG500-216 32GB RAM: corsair ddr4 3200mhz 32gb PSU: CGR GDN-850 SSD: Fanxiang S501 512GB 総額25万円弱でした なお、メモリは去年買ったので今この構成作るともっと高いかも。
Ryu
@_ryu15_Sep 15
ATX工業フレームマニュア
64
127
29
1,281
498,358
Ryu retweeted
Claude
@claudeai10h
Starting next week, sessions across Claude web, desktop, and mobile apps will run in auto mode by default on Pro, Max, and Team plans. Rather than asking you to approve each step, an automated check blocks risky actions. Switch back to manual anytime.
ClaudeDevs
@ClaudeDevsAug 7
Starting August 14, auto mode will be the default permission mode in Claude Code for Pro, Max, and Team users. Auto mode reviews shell commands and actions with a separate classifier. In testing, it caught 89% of dangerous commands. Manual approval caught 14%.
551
524
276
14,377
1,202,566
今月リセット、3つめ燃やしてしまいました; ;
15
418
Ryu retweeted
Tibo
@thsottiaux1h
Landing tomorrow before 10am PST together with a small ship.
193
53
111
1,231
158,523
Ryu retweeted
Arena.ai
@arena1h
A new model from @Kimi_Moonshot is coming. Stay tuned for where it lands in the Arena.
Arena.ai
@arenaJul 16
Big news: Kimi-K3 by @Kimi_Moonshot is now #1 in the Frontend Code Arena with 1679 pts, surpassing Claude Fable 5. This is a 17-place jump from Kimi-k2.6 (#18 -> #1). In Frontend, Kimi-K3 ranked #1 in 6 of 7 domains: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Simulations, and Content Creation Tools, landing #2 only in Gaming behind Fable 5. The full model weights will be released by July 27. Congrats to the @Kimi_Moonshot team on this major milestone!
26
23
5
485
24,191
Ryu retweeted
Tibo
@thsottiaux1h
Day 18/ Decisions API is GA, OpenResponses joined the Linux Foundation, we shipped some performance improvements and Uber is going all-in on codex & work. But yes… that’s a reset!
644
205
347
4,062
329,036
Command Code GOAT契約しました
2
21
731
うちは無検閲Qwen FlashNextをオーケストレーターにworkerをDS V4.1 Flash、GLM5.3 Flashで組んでるけどマジで何も問題ないしなんでもやってくれる ちな乞食仕様なのでマシン代33万と電気代以外かかってないで無料API駆使しまくっとる
inoino@inoino655367h
あまり知られてないけど、オーケストレーターモデルは単に一番強いモデルである必要は無いんだよなあ 無検閲GLM5.3flashにAstra、Opus5.5をぶん回させているけど全く問題ないで
2
26
1
333
14,797
時代はPWAなんよ なぜ流行らない A:制約が多いから
1
1
1
21
980
www
Ryu
@_ryu15_2h
ガチ草
1
1
238
あ、Alexandr Wang まずい
Alexandr Wang
@alexandr_wang4h
this is actually your first decision when downloading muse in japan
15
712
this is actually your first decision when downloading muse in japan
Muse
@Muse7h
こんにちは日本！ I came to eat sashimi and assist you in your everyday tasks... and I’m all out of sashimi. download here: muse.ai 🇯🇵
121
91
56
1,959
84,239
MiniMax M3.1 Flash weights are now public. 428B parameters total, 23B activated per token. Agentic performance: A generational leap in agentic tasks, coding, and cybersecurity — 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and 82.2 on CyberGym. Efficient by design: MiniMax Sparse Attention and DSpark speculative decoding improve inference efficiency. Hugging Face: huggingface.co/MiniMaxAI/Min… ModelScope: modelscope.cn/models/MiniMax… API: platform.minimax.io M Plan: minimax.io/m-planMiniMax MCode for desktop: agent.minimax.io/download

MiniMaxAI/MiniMax-M3.1-Flash · Hugging Face

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co
MiniMax (official)
@MiniMax_AIOct 19
MiniMax M3.1 Flash: Space Performance + Bunny Speed 🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash. 💰Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence across key tasks such as Cowork (Automation-Bench: 62%; ALE: 30%), Coding (DeepSWE: 74%) and Visual Reasoning (BabyVision: 91%), at 1/3 the cost of M3 and 3x the speed. API pricing (flat rate): $0.15 / $0.60 per 1M input/output tokens, with cache reads at $0.015 / 1M tokens. Model weights will arrive on Oct 23. API: platform.minimax.io M Plan: minimax.io/m-plan MiniMax Code for desktop: agent.minimax.io/download
87
158
99
1,492
103,023
ガチ草
Ryu
@_ryu15_4h
学習が1エポック終わった！ v100マシンの容量無くなってSSD買うまではモデル置けなくてさらに作業するのは厳しそうだったのでmodalへ() 今からしゃべってみるかー
1
2
1
62
1,933
Kimiくる！！！！！！！！！ KimiのPVが一番気持ちいいんだよなぁ...！
Kimi.ai
@Kimi_Moonshot4h
2
4
1
86
2,912
444
747
690
10,206
684,838
M3.1 Flash weights are now open 🔓 Thanks to @togethercompute for being a day-0 launch partner and bringing M3.1 Flash to production on the AI Native Cloud. Open weights, no GPUs to manage. Great building together! 🐰
Together AI
@togethercompute4h
minimax m3.1 flash is now live on together ai we’re excited to be a day-0 launch partner for the model, aka space bunny m3.1 flash is built for agentic tasks, coding, computer use + cybersecurity, with a 1m context window and vision support try it now together.ai/models/minimax-m…
17
6
8
109
6,445
学習が1エポック終わった！ v100マシンの容量無くなってSSD買うまではモデル置けなくてさらに作業するのは厳しそうだったのでmodalへ() 今からしゃべってみるかー
1
2
2
44
3,374
DeepSeek v4.1 Flash 1017でこれか... PROはどうなっちゃうんだ....？
きしだൠ(K1S)
@kis21h
DeepSeek-V4.1-Flash-1017、進化したなぁ
5
1
48
2,678
DeepSeek-V4.1-Flash-1017、進化したなぁ
3
10
1
69
5,665
Load more