diff --git a/CHANGELOG.md b/CHANGELOG.md
index 2e08ffd1..4e520b3e 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -12,9 +12,15 @@ and this project will adhere to [Semantic Versioning](https://semver.org/spec/v2
 - `/accounts/{stake_address}/addresses/assets` endpoint to list all assets on addresses related to a given stake_address (account)
 - `/epochs/latest/parameters` and `/blocks/latest/txs` endpoints to list the current information about latest epoch and block
 
+### Changed
+
+- implementation (increase allowed burst) of rate limiting and clarified documentation
+
 ### Fixed
 
 - `onchain_metadata` in `/assets/{asset}/total` endpoint that could have been missing in some special cases
+- `onchain_metadata` in `/assets/:asset` endpoint to return null on malformed data which do not follow https://github.com/cardano-foundation/CIPs/pull/85/files
+
 
 ## [0.1.15] - 2021-05-19
 
diff --git a/swagger.yaml b/swagger.yaml
index dbb7037e..c5b778e6 100644
--- a/swagger.yaml
+++ b/swagger.yaml
@@ -74,9 +74,14 @@ info:
     ```
     ## Limits
 
-     There are two types of limits we are enforcing. The first depends on your plan and is the number of request we allow per day. We defined
-     the day from midnight to midnight of UTC time. The second is rate limiting, where we limit an end user to 10 requests per second after a period
-     of 5 second burst. We believe this should be sufficient for most of the use cases. If it is not and you have a specific use case, please get in touch with us, and
+     There are two types of limits we are enforcing:
+
+     The first depends on your plan and is the number of request we allow per day. We defined the day from midnight to midnight of UTC time.
+
+     The second is rate limiting. We limit an end user, distinguished by IP address, to 10 requests per second. On top of that, we allow
+     each user to send burst of 500 requests, which cools off at rate of 10 requests per second. In essence, a user is allowed to make another
+     whole burst after (currently) 500/10 = 50 seconds. E.g. if a user attemtps to make a call 3 seconds after whole burst, 30 requests will be processed.
+     We believe this should be sufficient for most of the use cases. If it is not and you have a specific use case, please get in touch with us, and
      we will make sure to take it into account as much as we can.
 
 servers: