estia good first issues (Rust)

Use one wording for the fingerprint-mismatch error on both embedding routes 3 days ago

good first issue area: server

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

The llama adapter's hello should name the llama.cpp build 26 days ago

good first issue area: llama.cpp

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

A chat-template refusal from llama-server should be a 400, not a 500 27 days ago

good first issue area: llama.cpp area: server

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Benchmarks: llama.cpp against MLX on the same Mac, plus x86 CPU and NVIDIA rows 28 days ago

help wanted roadmap area: llama.cpp

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Idea: fuzz the protocol and HTTP parsers 28 days ago

help wanted idea area: security

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

OpenAI parity: `stop` sequences and `tool_choice` 28 days ago

enhancement help wanted area: server

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Access log: report the runner's decode rate 28 days ago

good first issue area: server

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Show memory per loaded model in `estia dashboard` 28 days ago

good first issue area: cli

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Refuse image parts with a 400 until image input is supported 28 days ago

good first issue area: server

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

`estia service install` should carry ESTIA_LLAMA_SERVER and ESTIA_LLAMA_ARGS into the service 28 days ago

good first issue area: llama.cpp area: cli

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Help text for the `<ID>` argument of `estia pull` 28 days ago

documentation good first issue area: cli

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

`estia models`: long model ids push the rest of the row out of line 28 days ago

good first issue area: cli

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Packages: a Homebrew tap, a .deb, a Windows installer 28 days ago

help wanted roadmap area: packaging

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

MLX: embeddings without mlx-embeddings (GPL-3.0) 28 days ago

help wanted roadmap area: mlx

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Thin SDKs for Python and TypeScript; typed errors in RemoteEngine; publish the crates 28 days ago

help wanted roadmap area: clients

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

A conformance suite for runners 28 days ago

help wanted roadmap area: engine

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

Release archives for Linux (x64, arm64) and Intel macOS 28 days ago

help wanted roadmap area: packaging platform: linux

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

llama.cpp: CUDA and ROCm variant choice, with a fallback when no GPU is listed 28 days ago

help wanted roadmap needs hardware area: llama.cpp platform: linux

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

llama.cpp: run Estia on Linux outside CI and report 28 days ago

help wanted roadmap needs hardware area: llama.cpp platform: linux

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

llama.cpp: run Gemma 4 GGUF end to end (tools, JSON Schema, embeddings) 28 days ago

help wanted roadmap area: llama.cpp

modelcaddy/estia

0

Local LLM inference engine: OpenAI-compatible server, MLX on Apple Silicon and llama.cpp everywhere, LAN pairing. Apache-2.0.

Rust

#gemma#inference#llama-cpp#llm#local-first#mlx#openai-api#rust

modelcaddy/estia

Open Issues Need Help