16 engineers on Hacker News describe using Llama 2 first-hand: 9 endorse it, 0 regret it, 1 are mixed.
“Thanks to AI fixing bugs in llama my inference runs 10x faster in worst case scenario.”
endorse · personal · Roark66 · 2026-10-23 · source ↗ · wrong label?“They are incredibly fast and cost-effective for specific, repetitive tasks.”
endorse · production · PaulShin · 2025-06-09 · source ↗ · wrong label?“we saw good performance with llama over our previous use of Mistral/Claude/OpenAI”
endorse · production · ivorbuk · 2024-09-24 · source ↗ · wrong label?“We've started using meta.ai and the 200B llama model on a couple servers we run at my company.”
endorse · production · sfblah · 2024-08-28 · source ↗ · wrong label?“We switched from text parsing and avg bad results to just using llama3 (with a little bit of saveguarding) and its a lot better.”
endorse · production · Bumblonono · 2024-07-21 · source ↗ · wrong label?“Groq with llama3 70b is so fast and good enough for what we do (source code stuff)”
endorse · production · anonzzzies · 2024-04-23 · source ↗ · wrong label?“it's actually good enough for llama-2-7b, insofar as it transforms this model from 'terrible with my prompt in production today'”
mixed · production · phillipcarter · 2023-12-13 · source ↗ · wrong label?“That's why we use Llama 2 70b over Falcon-180b, OPT-175b, and Bloom-175b.”
endorse · production · nabakin · 2023-12-09 · source ↗ · wrong label?“where fine-tuning even the 7B Llama-2 model outperforms GPT-4 (surprisingly)”
endorse · personal · robertnishihara · 2023-08-23 · source ↗ · wrong label?“We ditched most of our focus on Falcon 40B after Llama 2 70B came out, both the tokens per sec and quality of results are not even close.”
endorse · production · paulcjh · 2023-08-01 · source ↗ · wrong label?“our Llama 8b (via Groq) implementat”
neutral · production · minotauris · 2026-02-15 · source ↗ · wrong label?“I use llama 3B in prod for summarization (kinda).”
neutral · production · kianN · 2025-01-22 · source ↗ · wrong label?“for Llama 3.1 we found some show-stopping context length related bugs”
neutral · production · reissbaker · 2024-07-28 · source ↗ · wrong label?“If the model isn't hosted somewhere else, e.g. a lot of the Llama 3 finetunes, we run the models on our own GPU clusters”
neutral · production · reissbaker · 2024-07-24 · source ↗ · wrong label?“I've deployed the OpenAI-compatible completions API over Llama2 in production with vLLM, and it works perfectly.”
neutral · production · sbrother · 2023-11-18 · source ↗ · wrong label?“benchmarks we've run at Anyscale support this conclusion”
neutral · production · robertnishihara · 2023-08-13 · source ↗ · wrong label?