on:glm-5-3-flash

· 55 artifacts, sorted by favorites. · open in search — combine tags, sort, filter by date →

@Teknium 2026-08-21 — Ox Alpha now available in Hermes Agent through @opencode and @OpenRouter! ♥1225
@Teknium 2026-08-26 — Ox Alpha free period is over, but GLM-5.3 Flash is now available if you'd like to continue using it! https://t.co/7tqjDe ♥748
@Teknium 2026-09-20 — GLM-5.3 FlashX is now available in Hermes Agent through Nous Portal and OpenRouter! https://t.co/yoSljONuhh ♥530
@DanielleFong 2026-08-25 — ox alpha is 100% second order distilled from a claude. good god man ♥137
@__venki__ 2026-09-17 — Have a task that wants cheap text processing, and I’m trying all the cheap models. (Glm 5.3 flash, qwen 3.8 flash) They ♥81
@aliceisplaying 2026-09-01 — just ran this thru fable 5.1 with no tools. very sad that soon this will be in the training data also tried glm 5.3 fla ♥71
@aliceisplaying 2026-09-30 — i still cant believe anthropic wrote a promo blogpost for glm-5.3-flash-abliterated ♥48
@DanielleFong 2026-08-25 — do not mention ox-alpha to claude https://t.co/NMvJSZB2A2 ♥46
@NousResearch 2026-10-09 — Slight Correction: Step 5 Preview's final score on Hermes Index has been updated to 32.83, slightly behind GPT-6 Luna (3 ♥43
@liminal_bardo 2026-09-10 — Anthropic's claim that Moonshot were routing Kimi calls to Claude makes me think the same was happening with Ox Alpha / ♥33
@liminal_bardo 2026-08-21 — Ox Alpha stealth model on the only benchmark that matters. If it really is a Chinese model then it's super interesting. ♥31
@xlr8harder 2026-08-21 — Speechmap word use frequency on ox alpha is less helpful to identify it than you might hope, since everyone is copying a ♥30
@aliceisplaying 2026-09-22 — do not set glm-5.3-flash to max thinking, that model can overthink more than i do ♥29
@multimodalart 2026-08-26 — Ox Alpha is here 🔥 highlight: GLM-5.3-Flash starts from a new base model. architecture and training recipe redesigned. ♥27
@jmbollenbacher 2026-08-27 — I still wanna know how Luna is so token efficient compared to its peers (dsv4-flash, glm5.3-flash, etc.). It uses like ♥20
@jessi_cata 2026-10-06 — based on a comment by GLM-5.3-Flash on https://t.co/LrtR3PzhyC which features eccentric AI diction https://t.co/Zl0NyRA ♥11
@__venki__ 2026-09-18 — Conclusion from my experiments: Tested: Deepseek V4.1 flash, Qwen 3.8 flash, glm 5.3 flash, muse spark, gpt Luna Rough ♥8
@MikePFrank 2026-09-15 — Amazingly, according to Artificial Analysis, good old Ox Alpha (GLM-5.3 Flash), which I'm currently deploying in https:/ ♥8
@QiaochuYuan 2026-09-10 — @liminal_bardo i had a 2-turn conversation with ox-alpha, first turn sounded like GLM-5.3 and second turn sounded like c ♥7
@MikePFrank 2026-08-30 — Working by hand like a caveman instead of vibe-coding (since this is a DIY hobby project), I got image generation workin ♥7
@MikePFrank 2026-09-23 — Over the last couple of weeks since porting my Astra chess site over to use GLM-5.3 Flash instead to drive gameplay to r ♥6
@stevekrouse 2026-09-01 — So far, I much prefer deepseek v4 flash 0731 than glm 5.3 flash for building vals on val town. It's faster, cheaper, and ♥6
@jd_pressman 2026-08-21 — @teortaxesTex Well this answer is already better than Claude: User: Who are your favorite ethical philosophers? Ox-Alp ♥6
@christophcsmith 2026-09-03 — @Aron_Adler Cursor is down, GLM-5.3-Flash on OpenCode is down, I hear ChatGPT is down. The Singularity is canceled, eve ♥5
@UnderwaterBepis 2026-09-17 — Phase 1 of this plan to get Kimi K3 running at home us done, I can run 30 tok/sec GLM 5.3 Flash or DS 4.1 Flash (80-90 t ♥4
@jmbollenbacher 2026-08-26 — After seeing the architectures glm-5.3-flash and qwen-38-flash-next, i think maybe this take was wrong. We're going to ♥4
@MikePFrank 2026-08-24 — Trying out Ox Alpha under Hermes Agent hosted on the Nous Research cloud. Asked the model to search the web to learn a l ♥4
@geogristle 2026-09-09 — @aman_gif that's how much i pay for glm-5.3-flash ♥3
@MikePFrank 2026-08-27 — Updated my Ox Alpha bot to use GLM-5.3 Flash (misspelled in start message, derp) https://t.co/qjpBueo6RH ♥3
@jmbollenbacher 2026-08-26 — glm-5.3-flash and qwen3.8-flash-next in the same day. what an embarrassment of riches. ♥3
@billyhumblebrag 2026-08-22 — Ox Alpha is clearly a google model. The Phoenician letter aleph is a pictogram of an ox's head and literally means ox, t ♥3
@Nymne 2026-10-06 — @bbishdotdev @thdxr Get the model to outsource all the testing to a second completely different model. GLM5.3-Flash does ♥2
@MikePFrank 2026-09-14 — In case anyone wants to try playing against Ox Alpha in the chess harness built out earlier by Astra, it's now available ♥2
@MikePFrank 2026-09-13 — "Arcturus" (a.k.a. GLM-5.3 Flash, a.k.a. Ox Alpha) is now running within the chess harness built earlier by Astra for it ♥2
@QiaochuYuan 2026-09-10 — @liminal_bardo *GLM-5.2 ♥2
@geogristle 2026-09-09 — @guerre_doctrine 5090 asus rog astral and we got maybe rtx pro 6000 too can, use. i also have 10% normal price glm-5.3-f ♥2
@MikePFrank 2026-09-01 — Playing with GLM-5.3 Flash, I’ve been impressed with how level-headed it is, and how good it is at keeping track of mult ♥2
@qwweryo 2026-08-29 — GLM-5.3-Flash is the hot new open source model (Ox Alpha) and seems to be better than Opus 4.8 Unsloth shipped a 3-bit ♥2
@multimodalart 2026-08-26 — it's great that with Qwen3.8-Flash-Next and GLM-5.3-Flash, we got not only 2 high quality models, but also 2 novel model ♥2
@RobertHaisfield 2026-09-26 — @Nikos1337 GLM 5.3 and K3 are not in the same price range. GLM 5.3 Flash and Deepseek V4.1 Flash are in the same price r ♥1
@johnhanacek 2026-09-14 — @jordanurbs GLM5.3 Flash and DeepSeek 4.1 flash are the wind beneath my wings, Only a Chinese fable can lower the rent. ♥1
@johnhanacek 2026-09-10 — Me when I'm working on three simultaneous projects but hit the Fable 5.1 weekly limit and don't have budget for Astra ri ♥1
@MikePFrank 2026-09-10 — Arcturus, my Ox Alpha bot, is jelly that Astra got to build out a whole public-facing chess app for itself while Arky’s ♥1
@geogristle 2026-09-09 — @k7agar how does e.g. glm-5.3-flash or qwen3.8 max do on this ♥1
@MikePFrank 2026-08-30 — Some more art prompted from gpt-image-2 by Arcturus bot (the model formerly known as Ox Alpha). https://t.co/vx9MjTBo0G ♥1
@Coobyk_ 2026-08-24 — @blueemi99 Currently Ox Alpha cuz it's free, DeepSeek V4 Flash cuz it's so cheap, and Muse Spark is also really cheap on ♥1
@lorde_russell 2026-08-22 — @Linahuaa great model but Ox Alpha is NOT fable class be serious ♥1
@januarycomputer 2026-08-22 — @wolfie_ nah, not really. kimi k2.5 vs composer 2.5. in fact, the refusals you hit with Ox Alpha are on the provider sid ♥1
@LinXule 2026-08-21 — Ox Alpha on its lost worthy rivals (GPT, Claude, Gemini, DeepSeek, Llama, Qwen, Mistral). It’s Claude-shaped but also wi ♥1
@jitl 2026-10-04 — @fedor @felipecsl many open models still get really dumb at >half context full. GLM 5.3 Flash and Deepseek 4.1 Flash i’v ♥0
@MikePFrank 2026-10-03 — Someone shared what I thought was a pretty interesting game against my Ox Alpha (GLM-5.3 Flash) powered chess bot at htt ♥0
@jmbollenbacher 2026-08-28 — @LandonExplr models are already useful at small sizes, and theyre getting better every month. see glm-5.3-flash and qwe ♥0
@whitenoisemachn 2026-08-26 — okay now that Ox Alpha is confirmed to be GLM-5.3-Flash, what was up with the Google folks vagueposting? Gemini 4 better ♥0
@FoldingNapkins 2026-08-25 — Will Accelerated Understanding hit with a bang or a fizzle? Is the mysterious newcomer "Ox Alpha" going to show their me ♥0
@csgbwk 2026-08-22 — There's a non zero chance Ox Alpha is both gemini and a benchmaxxed GLM 5.2 finetune ♥0