simdjson
No written verdict yet. Here is what commenters said.
5recommend
1warn against
Nov 12, 2025first mention
Jul 23, 2026latest mention
Why people recommend it
“It showcases an especially ingenious scheme for escaping json strings”
“You could just take simdjson, use its ondemand api”
Show 2 more
“can process gigabytes per second”
“You can parse JSON at several GB/s: https://github.com/simdjson/simdjson”
Caveats and dissent
“any usage of it breaks std::unordered_map in unrelated parts of the code”
Who recommends it in their own words
Threads where it came up
- Command-line Tools can be 235x Faster than your Hadoop Cluster (2014) 2 mentions
- GigaToken: ~1000x faster Language model tokenization 1 mention
- A Faster Alternative to Jq 1 mention
- We rewrote our Rust WASM Parser in TypeScript – and it got 3x Faster 1 mention
- Ask HN: How do you budget for token based AI APIs? 1 mention
- Show HN: I built a CSV parser to try Go 1.26's new SIMD package 1 mention
Also written as simd-json. Data: simdjson.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 11
“It showcases an especially ingenious scheme for escaping json strings”
“You could just take simdjson, use its ondemand api”
“can process gigabytes per second”
“Parsing the session logs directly (JSONL/JSON) with simd-json”
“Simdjson uses this trick, and I use it as well”
> https://github.com/simdjson/simdjson Was not aware of this but seems it is not there natively in Python,but seems cool. Will try out in future.
“You can parse JSON at several GB/s: https://github.com/simdjson/simdjson”
“is able to outperform simdjson by 120x in some benchmarks”
“any usage of it breaks std::unordered_map in unrelated parts of the code”
“multi-threaded parsing (simdjson)”
“Multi-threaded parsing with simdjson's on-demand API”