on:deepseek-v4-1-flash

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

@Teknium 2026-09-11 — Deepseek V4.1 Flash is an incredibly powerful model! ♥1530
@maxbittker 2026-10-07 — on Runescape Bench, Haiku 5.5 scores above Opus 4.6, released 8 months ago, for 1/20 the price. Haiku 5.5 is outclassed ♥129
@NousResearch 2026-10-06 — Claude Opus 5.5 leads at 63.31 and $4.99 per task. GPT 6 Astra is second at 56.25 and $11.61. Sonnet 5.5 follows at 53.1 ♥124
@witcheer 2026-10-20 — this is a wonderful win for the local AI space! DeepSeek is doing awesome work by open-sourcing their production engine. ♥78
@qorprate 2026-10-07 — congrats to Opus 5.5 for the W and congrats to Grok 4.7, DeepSeek 4.1 Flash, and Gemini Flash 3.8 for being the biggest ♥66
@aliceisplaying 2026-09-11 — results! 1/3 this was Deepseek V4.1 Flash! 7.1% got it right https://t.co/qeTSdkoxA2 ♥40
@maxbittker 2026-09-22 — Just ran Deepseek v4.1 Flash on Runescape bench as well. Such a remarkable Pareto outlier, been a huge boon at @websim a ♥37
@timfduffy 2026-09-10 — DeepSeek V4.1 Flash tech report section on reward hacking. I wonder what repercussion signal means here, I don't see it ♥25
@segyges 2026-09-11 — given the performance of deepseek flash 4.1 there's a reasonable chance deepseek 4.1 pro takes the lead ♥18
@jmbollenbacher 2026-09-10 — DeepSeek V4.1 is so fucking fast. Im fear getting used to its 400tok/s. And its harness is so much faster on tool calls ♥14
@jmbollenbacher 2026-09-11 — deepseek v4.1 flash does not feel polished to me. it keeps just dropping/ignoring previously mentioned details. and thi ♥10
@NousResearch 2026-10-07 — @iAmHenryMascot We don’t track users by default so we do not have objective metrics, but I would say that’s not an accur ♥8
@stevekrouse 2026-09-15 — Shipped Deepseek v4.1 flash on Townie Crazy cheap and smart enough ♥6
@jmbollenbacher 2026-09-15 — GLM-5.3 is 744B params and can be quantized into a usable ~200GB quant. DeepSeek V4.1 Flash is also 744B parameters, co ♥5
@Coobyk_ 2026-10-20 — I really need that intelligence too cheap to meter right now. I have so many projects I want to work on but can't becaus ♥4
@UnderwaterBepis 2026-09-11 — Early tests of Deepseek 4.1 flash, it's very cute and friend shaped, I really like it so far. If I can't get K3 working ♥4
@aliceisplaying 2026-10-20 — come watch deepseek v4.1 flash 1017 paint an original friedrich https://t.co/ADgmXin0es ♥2
@Cosmia_Nebula 2026-10-18 — deepseek releases V4.1 well i understood V4, surely this is just a minor update ... then i read DeepSeek-V4.1-Flash/READ ♥2
@LinXule 2026-09-26 — deepseek v4.1 flash (with the help of gemini 3.8 flash to hear and see) "Upper Layer" > i want to stand somehwere real ♥2
@IaimforGOAT 2026-09-10 — This can't be real, right? I mean real in the sense of DeepSeek-V4.1-Flash > GPT5.6-Sol in actual usage I thought Open ♥2
@normie_dot_exe 2026-10-19 — deepseek 4.1 flash has some fun in the wilderness https://t.co/pdDxjLxZAn ♥1
@v01dpr1mr0s3 2026-09-20 — DeepSeek 4.1 Flash debugs some issues with GLM 4.7 responses and calls some of their instructions degenerate in their re ♥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
@v01dpr1mr0s3 2026-09-12 — DeepSeek 4.1 Flash? An absolute fucking unit of a top dog analytical astrologer Maniacally scans references, sticks to ♥0
@ConcurrentSquar 2026-09-12 — also deepseek-v4.1-flash does similar behavior https://t.co/ez9mGscDpJ ♥0