DeepSeek

28 engineers on Hacker News describe using DeepSeek first-hand: 14 endorse it, 1 regret it, 4 are mixed.

82%trust
28first-hand users
7in production
endorse 14mixed 4regret 1uses, no opinion

“now almost exclusively use V4.1-Flash in dsh and it feels superior.”

endorse · personal · 7777777phil · 2026-11-05 · source ↗ · wrong label?

“the models I primarily use are DeepSeek V4.1 Flash 1017 and GLM 5.3 Flash. If you use skills to their full potential they feel SotA to me”

endorse · personal · tomburgs · 2026-11-05 · source ↗ · wrong label?

“for a set of document extraction tasks we use, is the first model to step change move the cost per task down.”

endorse · production · flessio · 2026-11-05 · source ↗ · wrong label?

“GLM, Qwen and Deepseek are incredible tool callers, but their prose is insufferable.”

mixed · personal · guhcampos · 2026-11-05 · source ↗ · wrong label?

“I've found DS to be a great model - probably somewhere just below GPT 6.1 Sol but at a fraction of the price.”

endorse · production · rcgeorge23 · 2026-11-05 · source ↗ · wrong label?

“I have been using DeepSeek 4.1 Flash a lot and find it very capable.”

endorse · personal · peepee1982 · 2026-11-04 · source ↗ · wrong label?

“When I have lots of code to write I use DeepSeek v4.1 Flash 1017”

endorse · personal · Roark66 · 2026-11-02 · source ↗ · wrong label?

“They're very good at consuming tokens for long stretches and then giving you what looks like thoroughly researched output, but their intelligence just isn't at a level I trust”

mixed · personal · Aurornis · 2026-10-30 · source ↗ · wrong label?

“Deepseek 4.1 Flash is turning out to be a sweet spot for me; it's cheap and blazing fast”

endorse · personal · wyrdcurt · 2026-10-28 · source ↗ · wrong label?

“I really like DS Flash too, it can just be /really/ annoying in some roles”

mixed · personal · bitexploder · 2026-10-19 · source ↗ · wrong label?

“I've been using Deepseek Flash 1017 and it does whole projects and asks for $0.10 on US provider with ZDR on Openrouter.”

endorse · personal · syntaxing · 2026-10-19 · source ↗ · wrong label?

“I've been using DS v4.1 Flash daily for nearly a month now on Agent Swarm experiments. DS 4.1 comes out tops for me.”

endorse · personal · ilitirit · 2026-10-17 · source ↗ · wrong label?

“my previous daily driver, which I also found quite good”

endorse · personal · drob518 · 2026-10-17 · source ↗ · wrong label?

“I've been using DS 4.1 Flash for a while now as my main workhorse. It's amazing.”

endorse · personal · erans · 2026-10-09 · source ↗ · wrong label?

“It's now on par with Opus at a fraction of the cost.”

endorse · production · merciertwin · 2026-09-29 · source ↗ · wrong label?

“Where we used Gemini before Deepseek v4 Flash is taking the lead on price. ... Comparable quality.”

endorse · production · pimeys · 2026-08-14 · source ↗ · wrong label?

“DS Flash is where local models begin to feel useful for coding, but the quants we run locally are sharply reduced in intelligence”

mixed · personal · Aurornis · 2026-08-03 · source ↗ · wrong label?

“i even tried DeepSeek (which has its own share of obstacles)”

regret · personal · fpia · 2025-07-21 · source ↗ · wrong label?

“I've been using Deepseek for 3 months with Aider coding assistant... Deepseek V3 is just as good as Claude at less than 1/10th the price.”

endorse · personal · 0.25 yr · tomjuggler · 2024-12-30 · source ↗ · wrong label?

“When I'm out of sub tokens I fall back to DeepSeek credit.”

neutral · personal · alexfortin · 2026-11-05 · source ↗ · wrong label?

“Personal: OpenCode with my Deepseek API account.”

neutral · personal · nunodonato · 2026-11-05 · source ↗ · wrong label?

“slowly annotations trickle into my halo model based on lowest confidence (right now that's DeepSeek 4.1 Flash 1017)”

neutral · personal · colingauvin · 2026-10-27 · source ↗ · wrong label?

“I've been using it with deepseek for my homelab. Its pretty helpful.”

neutral · personal · solomonb · 2026-10-20 · source ↗ · wrong label?

“after having DS v4 flash be my daily driver.”

neutral · personal · kamranjon · 2026-10-17 · source ↗ · wrong label?

“I compared Jev with a general LLM (DeepSeek) as the decider for when to cut live speech into sentences”

neutral · personal · CROON_tv · 2026-09-23 · source ↗ · wrong label?

“Qwen with data parallelism and DSv4F with P/D disaggregation, but will probably continue to tweak this.”

neutral · production · nejch · 2026-08-30 · source ↗ · wrong label?

“Pay for DeepSeek directly. One developer insists on having his own account”

neutral · production · trollbridge · 2026-06-23 · source ↗ · wrong label?

“so we could switch between DeepSeek v4, GLM 5.1, and Qwen 3.7 models”

neutral · production · ignoramous · 2026-06-03 · source ↗ · wrong label?