mention how fast geocode is

README.md

@@ -49,7 +49,7 @@ See [FAQ](https://github.com/jqnatividad/qsv/discussions/categories/faq) for mor
 | [fmt](/src/cmd/fmt.rs#L2) | Reformat a CSV with different delimiters, record terminators or quoting rules. (Supports ASCII delimited data.)  |
 | [foreach](/src/cmd/foreach.rs#L3)<br>✨ | Loop over a CSV to execute shell commands. (not available on Windows)  |
 | [frequency](/src/cmd/frequency.rs#L2)<br>📇😣🏎️ | Build [frequency tables](https://statisticsbyjim.com/basics/frequency-table/) of each column. Uses multithreading to go faster if an index is present. |
-| [geocode](/src/cmd/geocode.rs#L2)<br>✨🧠🌐🚀 | Geocodes a location against an updatable local copy of the [Geonames](https://www.geonames.org/) cities database.  |
+| [geocode](/src/cmd/geocode.rs#L2)<br>✨🧠🌐🚀 | Geocodes a location against an updatable local copy of the [Geonames](https://www.geonames.org/) cities database. With caching and multi-threading, it geocodes up to 360,000 records/sec! |
 | [generate](/src/cmd/generate.rs#L2)<br>✨ | Generate test data by profiling a CSV using [Markov decision process](https://crates.io/crates/test-data-generation) machine learning.  |
 | [headers](/src/cmd/headers.rs#L2) | Show the headers of a CSV. Or show the intersection of all headers between many CSV files. |
 | [index](/src/cmd/index.rs#L2) | Create an index (📇) for a CSV. This is very quick (even the 15gb, 28m row NYC 311 dataset takes all of 15 seconds to index) & provides constant time indexing/random access into the CSV. With an index, `count`, `sample` & `slice` work instantaneously; random access mode is enabled in `luau`; and multithreading (🏎️) is enabled for the `frequency`, `split`, `stats`, `schema` & `tojsonl` commands. |