tkmind/tkmind_go
1
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity

feat: add local inference provider with llama.cpp backend and HuggingFace model management (#6933)

Browse Source 
Co-authored-by: Douwe Osinga <douwe@squareup.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: jh-block <jhugo@block.xyz>
Co-authored-by: Spence <spencermartin@squareup.com>
Co-authored-by: Michael Neale <michael.neale@gmail.com>
This commit is contained in:
Douwe Osinga
2026-02-19 18:30:05 +00:00
committed by GitHub
parent 6928c8cee1
commit ddd35f6d47
44 changed files with 7171 additions and 181 deletions
Download Patch File Download Diff File Expand all files Collapse all files
crates/goose/Cargo.toml
+6 -2
View File
@@ -9,7 +9,7 @@ description.workspace = true
[features]
default = []
cuda = ["candle-core/cuda", "candle-nn/cuda"]
cuda = ["candle-core/cuda", "candle-nn/cuda", "llama-cpp-2/cuda"]
[lints]
workspace = true
@@ -126,14 +126,17 @@ ignore = { workspace = true }
which = { workspace = true }
pctx_code_mode = "^0.2.3"
unbinder = "0.1.7"
llama-cpp-2 = { version = "0.1.133", features = ["sampler"] }
encoding_rs = "0.8.35"
[target.'cfg(target_os = "windows")'.dependencies]
winapi = { version = "0.3", features = ["wincred"] }
# Platform-specific GPU acceleration for Whisper
# Platform-specific GPU acceleration for Whisper and local inference
[target.'cfg(target_os = "macos")'.dependencies]
candle-core = { version = "0.9", default-features = false, features = ["metal"] }
candle-nn = { version = "0.9", default-features = false, features = ["metal"] }
llama-cpp-2 = { version = "0.1.133", features = ["sampler", "metal"] }
[dev-dependencies]
serial_test = { workspace = true }
@@ -156,6 +159,7 @@ path = "examples/agent.rs"
name = "databricks_oauth"
path = "examples/databricks_oauth.rs"
[[bin]]
name = "build_canonical_models"
path = "src/providers/canonical/build_canonical_models.rs"