diff --git a/Cargo.lock b/Cargo.lock
index 69b9599a..aa6d27e0 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -3143,7 +3143,7 @@ dependencies = [
 
 [[package]]
 name = "tw-api"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "serde",
  "serde_json",
@@ -3154,7 +3154,7 @@ dependencies = [
 
 [[package]]
 name = "tw-bedrock"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "aws-credential-types",
  "aws-sigv4",
@@ -3175,7 +3175,7 @@ dependencies = [
 
 [[package]]
 name = "tw-breaker"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "serde",
  "serde_json",
@@ -3183,7 +3183,7 @@ dependencies = [
 
 [[package]]
 name = "tw-config"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "blake3",
  "libc",
@@ -3208,7 +3208,7 @@ dependencies = [
 
 [[package]]
 name = "tw-control"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "axum",
  "base64",
@@ -3251,7 +3251,7 @@ dependencies = [
 
 [[package]]
 name = "tw-dialect"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "memchr",
  "serde",
@@ -3260,7 +3260,7 @@ dependencies = [
 
 [[package]]
 name = "tw-engine"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "chrono",
  "serde",
@@ -3274,7 +3274,7 @@ dependencies = [
 
 [[package]]
 name = "tw-gateway"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "arc-swap",
  "async-stream",
@@ -3325,7 +3325,7 @@ dependencies = [
 
 [[package]]
 name = "tw-guard"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "aho-corasick",
  "base64",
@@ -3343,7 +3343,7 @@ dependencies = [
 
 [[package]]
 name = "tw-link"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "serde",
  "serde_json",
@@ -3357,7 +3357,7 @@ dependencies = [
 
 [[package]]
 name = "tw-observe"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "tokio",
  "tracing",
@@ -3366,7 +3366,7 @@ dependencies = [
 
 [[package]]
 name = "tw-plugin"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "libc",
  "rand 0.10.2",
@@ -3381,7 +3381,7 @@ dependencies = [
 
 [[package]]
 name = "tw-pricing"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "arc-swap",
  "flate2",
@@ -3394,14 +3394,14 @@ dependencies = [
 
 [[package]]
 name = "tw-secret"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "thiserror",
 ]
 
 [[package]]
 name = "tw-store"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "blake3",
  "bytes",
@@ -3421,7 +3421,7 @@ dependencies = [
 
 [[package]]
 name = "tw-types"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "serde",
  "serde_json",
@@ -3430,7 +3430,7 @@ dependencies = [
 
 [[package]]
 name = "tw-watch"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "libc",
  "notify",
@@ -3441,7 +3441,7 @@ dependencies = [
 
 [[package]]
 name = "tw-yaml"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "saphyr-parser",
  "serde_yaml_ng",
@@ -3451,7 +3451,7 @@ dependencies = [
 
 [[package]]
 name = "twcore"
-version = "0.75.0"
+version = "0.76.0"
 dependencies = [
  "anyhow",
  "axum",
diff --git a/Cargo.toml b/Cargo.toml
index 436260ae..8d4796e5 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -54,7 +54,7 @@ categories = ["network-programming", "web-programming::http-server"]
 
 # 二进制走 CalVer，crate 走 SemVer —— 两者是两回事，卖给不同的人。
 # 这里是 crate 的版本。
-version = "0.75.0"
+version = "0.76.0"
 
 [workspace.dependencies]
 # ── 内部 crate ───────────────────────────────────────────
diff --git a/release-notes/0.76.0.md b/release-notes/0.76.0.md
new file mode 100644
index 00000000..6fc28ae4
--- /dev/null
+++ b/release-notes/0.76.0.md
@@ -0,0 +1,16 @@
+This release makes `/live` report how fast running requests are generating right now, so a client can show a live speed during a long request, and names a broken answer body instead of reporting a decoding error.
+
+**Upgrade notes**
+
+- The control-plane protocol version is now 49. ThinkWatch Lite must be upgraded together with core: a Lite release with protocol 48 does not connect to 0.76.0, and the other way round.
+- The request store's schema and the configuration are unchanged; nothing is cleared.
+- For crates that embed the gateway: `Ending::streaming` takes the gateway's `pace` table, `AppState` has a `pace` field, `tw_observe::Meter` and `EventBus::meter` are new, and `tw_engine::TextTokens` is public. The public API of `tw_guard` is unchanged.
+
+**Live generation speed.** `/live` used to give the speed of requests that finished in the last minute, so it had nothing to show while a long request ran.
+- The gateway keeps reading a successful stream after its first token and estimates the output as it arrives: text, reasoning text and tool arguments, with the same rule as the input estimate.
+- The estimate is multiplied by a ratio per upstream and model. Each streaming request that ends with reported usage updates that ratio (moving average, bounded between 0.5 and 2; skipped when reasoning was hidden). Gemini's cumulative usage in the stream is used when it is higher.
+- `LiveView.tokens_per_sec` is the total of the running requests over the last 3 seconds: null when nothing runs, 0 when requests run but none is generating yet.
+- `RunningView` has `tokens_per_sec` and `output_tokens` for each running request, null before the first token and for non-streaming requests.
+- The speed recorded for a finished request is unchanged and still uses the usage the upstream reports.
+
+**Broken answer bodies.** An upstream that hung up in the middle of an answer was reported as "error decoding response body". Such failures now have their own code, `gw.upstream.body_broken`, with the causes spelled out, and `gw.upstream.forward_failed` carries every cause, with the URL reduced to its origin.