From b1481b0e2eac737ef0eec8eeff84f838fc902555 Mon Sep 17 00:00:00 2001 From: Meng Zhang Date: Mon, 27 Nov 2023 14:57:45 +0800 Subject: [PATCH] chore: release 0.6.0 (#882) * add loadtest * release 0.6.0 * Release 0.6.0-rc.0 http-api-bindings@0.6.0-rc.0 juniper-axum@0.6.0-rc.0 llama-cpp-bindings@0.6.0-rc.0 tabby@0.6.0-rc.0 tabby-common@0.6.0-rc.0 tabby-download@0.6.0-rc.0 tabby-inference@0.6.0-rc.0 tabby-scheduler@0.6.0-rc.0 tabby-webserver@0.6.0-rc.0 Generated by cargo-workspaces * Release 0.6.0-rc.1 http-api-bindings@0.6.0-rc.1 juniper-axum@0.6.0-rc.1 llama-cpp-bindings@0.6.0-rc.1 tabby@0.6.0-rc.1 tabby-common@0.6.0-rc.1 tabby-download@0.6.0-rc.1 tabby-inference@0.6.0-rc.1 tabby-scheduler@0.6.0-rc.1 tabby-webserver@0.6.0-rc.1 Generated by cargo-workspaces * Release 0.6.0-rc.2 http-api-bindings@0.6.0-rc.2 juniper-axum@0.6.0-rc.2 llama-cpp-bindings@0.6.0-rc.2 tabby@0.6.0-rc.2 tabby-common@0.6.0-rc.2 tabby-download@0.6.0-rc.2 tabby-inference@0.6.0-rc.2 tabby-scheduler@0.6.0-rc.2 tabby-webserver@0.6.0-rc.2 Generated by cargo-workspaces * Release 0.6.0 http-api-bindings@0.6.0 juniper-axum@0.6.0 llama-cpp-bindings@0.6.0 tabby@0.6.0 tabby-common@0.6.0 tabby-download@0.6.0 tabby-inference@0.6.0 tabby-scheduler@0.6.0 tabby-webserver@0.6.0 Generated by cargo-workspaces --- CHANGELOG.md | 8 +++++++- Cargo.lock | 18 +++++++++--------- Cargo.toml | 2 +- crates/http-api-bindings/Cargo.toml | 2 +- crates/llama-cpp-bindings/Cargo.toml | 2 +- crates/tabby-common/Cargo.toml | 2 +- crates/tabby-download/Cargo.toml | 2 +- crates/tabby-inference/Cargo.toml | 2 +- crates/tabby-scheduler/Cargo.toml | 2 +- crates/tabby/Cargo.toml | 2 +- tests/default.loadtest.js | 7 ++----- website/docs/installation/modal/app.py | 4 +++- website/docs/installation/modal/index.md | 2 +- 13 files changed, 30 insertions(+), 25 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index dd392f6..489202b 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,4 +1,10 @@ -# v0.6.0 [Unreleased] +# v0.7.0 [Unreleased] + +## Features + +## Fixes and Improvements + +# v0.6.0 ## Features diff --git a/Cargo.lock b/Cargo.lock index 8e4dc9b..cb4bdaf 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -1951,7 +1951,7 @@ dependencies = [ [[package]] name = "http-api-bindings" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "async-trait", "futures", @@ -2254,7 +2254,7 @@ dependencies = [ [[package]] name = "juniper-axum" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "axum", "juniper", @@ -2410,7 +2410,7 @@ checksum = "da2479e8c062e40bf0066ffa0bc823de0a9368974af99c9f6df941d2c231e03f" [[package]] name = "llama-cpp-bindings" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "async-stream", "async-trait", @@ -4554,7 +4554,7 @@ dependencies = [ [[package]] name = "tabby" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "anyhow", "assert-json-diff", @@ -4604,7 +4604,7 @@ dependencies = [ [[package]] name = "tabby-common" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "anyhow", "async-trait", @@ -4624,7 +4624,7 @@ dependencies = [ [[package]] name = "tabby-download" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "aim", "anyhow", @@ -4636,7 +4636,7 @@ dependencies = [ [[package]] name = "tabby-inference" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "async-stream", "async-trait", @@ -4649,7 +4649,7 @@ dependencies = [ [[package]] name = "tabby-scheduler" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "anyhow", "cargo-lock", @@ -4680,7 +4680,7 @@ dependencies = [ [[package]] name = "tabby-webserver" -version = "0.6.0-dev" +version = "0.6.0" dependencies = [ "anyhow", "async-trait", diff --git a/Cargo.toml b/Cargo.toml index 236815e..d75e376 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -13,7 +13,7 @@ members = [ ] [workspace.package] -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" authors = ["Meng Zhang"] homepage = "https://github.com/TabbyML/tabby" diff --git a/crates/http-api-bindings/Cargo.toml b/crates/http-api-bindings/Cargo.toml index 2b8b3ed..fc2d344 100644 --- a/crates/http-api-bindings/Cargo.toml +++ b/crates/http-api-bindings/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "http-api-bindings" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" [dependencies] diff --git a/crates/llama-cpp-bindings/Cargo.toml b/crates/llama-cpp-bindings/Cargo.toml index 5dbf46f..c4840df 100644 --- a/crates/llama-cpp-bindings/Cargo.toml +++ b/crates/llama-cpp-bindings/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "llama-cpp-bindings" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" [features] diff --git a/crates/tabby-common/Cargo.toml b/crates/tabby-common/Cargo.toml index 755559d..f24ae02 100644 --- a/crates/tabby-common/Cargo.toml +++ b/crates/tabby-common/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "tabby-common" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" [dependencies] diff --git a/crates/tabby-download/Cargo.toml b/crates/tabby-download/Cargo.toml index 5e9eeaa..c4909a2 100644 --- a/crates/tabby-download/Cargo.toml +++ b/crates/tabby-download/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "tabby-download" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" [dependencies] diff --git a/crates/tabby-inference/Cargo.toml b/crates/tabby-inference/Cargo.toml index 4a2479b..509c513 100644 --- a/crates/tabby-inference/Cargo.toml +++ b/crates/tabby-inference/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "tabby-inference" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html diff --git a/crates/tabby-scheduler/Cargo.toml b/crates/tabby-scheduler/Cargo.toml index c8cbab7..7b1da75 100644 --- a/crates/tabby-scheduler/Cargo.toml +++ b/crates/tabby-scheduler/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "tabby-scheduler" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html diff --git a/crates/tabby/Cargo.toml b/crates/tabby/Cargo.toml index 2c804e3..4e4e504 100644 --- a/crates/tabby/Cargo.toml +++ b/crates/tabby/Cargo.toml @@ -1,6 +1,6 @@ [package] name = "tabby" -version = "0.6.0-dev" +version = "0.6.0" edition = "2021" [features] diff --git a/tests/default.loadtest.js b/tests/default.loadtest.js index 5ee183b..415d2eb 100644 --- a/tests/default.loadtest.js +++ b/tests/default.loadtest.js @@ -7,13 +7,10 @@ export const options = { { duration: "20s", target: 8 }, { duration: "5s", target: 0 }, ], - hosts: { - "api.tabbyml.com": __ENV.TABBY_API_HOST, - }, // Below thresholds are tested against TabbyML/StarCoder-1B served by NVIDIA T4 GPU. thresholds: { http_req_failed: ['rate<0.01'], // http errors should be less than 1% - http_req_duration: ["med<2500", "avg<2200", "p(95)<2500"], + http_req_duration: ["med<1800", "avg<1800", "p(95)<2000"], }, }; @@ -25,7 +22,7 @@ export default () => { }, }); const headers = { "Content-Type": "application/json" }; - const res = http.post("http://api.tabbyml.com/v1/completions", payload, { + const res = http.post(`${__ENV.TABBY_API_HOST}/v1/completions`, payload, { headers, }); check(res, { success: (r) => r.status === 200 }); diff --git a/website/docs/installation/modal/app.py b/website/docs/installation/modal/app.py index e6c4cdd..99286d8 100644 --- a/website/docs/installation/modal/app.py +++ b/website/docs/installation/modal/app.py @@ -4,7 +4,7 @@ modal serve app.py from modal import Image, Stub, asgi_app, gpu -IMAGE_NAME = "tabbyml/tabby:0.5.5" +IMAGE_NAME = "tabbyml/tabby:0.6.0" MODEL_ID = "TabbyML/StarCoder-1B" GPU_CONFIG = gpu.T4() @@ -58,6 +58,8 @@ def app(): "8000", "--device", "cuda", + "--parallelism", + "4", ] ) diff --git a/website/docs/installation/modal/index.md b/website/docs/installation/modal/index.md index 78f6f3b..86638d7 100644 --- a/website/docs/installation/modal/index.md +++ b/website/docs/installation/modal/index.md @@ -13,7 +13,7 @@ from modal import Image, Stub, asgi_app, gpu Next, we set the base docker image version, which model to serve, taking care to specify the GPU configuration required to fit the model into VRAM. ```python -IMAGE_NAME = "tabbyml/tabby:0.5.5" +IMAGE_NAME = "tabbyml/tabby:0.6.0" MODEL_ID = "TabbyML/StarCoder-1B" GPU_CONFIG = gpu.T4() ```