NOirBRight/dsh-llm-ollama

Ollama Cloud native chat adapter: direct NDJSON translation of the Ollama /api/chat wire format, model discovery with context windows and capabilities, and web search/fetch providers under the ollama-cloud id, plus a Web settings card.

Ollama Cloud integration for DeepSeek Harness. Chat uses Ollama's OpenAI-compatible Chat Completions endpoint through the shared pi-ai-backed adapter; model discovery and the Web Search/Fetch providers remain on Ollama-native APIs because those independent capabilities are not part of the chat protocol. The package root exposes the Cordis plugin contract and OllamaAdapter; the same artifact exports ./client, which contributes an Ollama Cloud card under Settings → LLM Providers. The protocol/capability split is recorded in ADR 0001. By default the settings RPC is loopback-only; remote browsers can't edit plugin settings unless configured.

Models & Providers ★ 3 updated 2026-08-27
View on GitHub ↗

Install

dsh plugin --profile web add --force https://github.com/NOirBRight/dsh-llm-ollama/releases/download/v0.6.24/dsh-llm-ollama-0.6.24.tgz

GitHub install per README (English): dsh plugin --profile web add --force https://github.com/NOirBRight/dsh-llm-ollama/releases/download/v0.6.24/dsh-llm-ollama-0.6.24.tgz, then dsh web. The README installs the companion UI package dsh-llm-providers-ui from its own release tarball in the same block (v0.2.9 as of 2026-09-18). Requires DeepSeek Harness 0.1.0-rc.6 or later. The repository tracks release-ready lib artifacts, so GitHub installation needs no build-script allowlist; a source checkout can use a link installation after pnpm run build. Re-verified 2026-09-18: README install block = two release tarballs pinned v0.2.9 (UI) and v0.6.24 (this plugin). Entry had the github: form pinned #v0.6.6 — two releases behind.

Compatibility

DeepSeek Harness 0.1.0-rc.6 or later. Ollama Cloud (OpenAI-compatible Chat Completions endpoint through the shared pi-ai-backed adapter).

Details

Recent updates

Ollama Cloud chat via pi-ai adapter; model discovery + search/fetch on Ollama-native APIs; Settings card; loopback-only RPC by default.

FAQ

Which DSH version is needed?
DeepSeek Harness 0.1.0-rc.6 or later.
How is chat routed?
Chat uses Ollama's OpenAI-compatible Chat Completions endpoint through the shared pi-ai-backed adapter; model discovery and search/fetch stay on Ollama-native APIs.
Can a remote browser configure it?
Not by default — the settings RPC is loopback-only unless explicitly configured.

Alternatives

Letter2025/dsh-model-failover · tianji-qingtian/dsh-model-router · JuneLearn/dsh-reasoning-settings

More plugins in Models & Providers

Browse more in Models & Providers

Guides for Models & Providers plugins