{"schemaVersion":1,"id":"ollama-local","revision":"2026-09-15.1","name":"Ollama (local runtime)","monogram":"O","category":"Packaged runtime","openness":"Open source","license":"MIT runtime source; model weights have separate licenses","color":"#f0b881","summary":"Local model runner and HTTP API, scoped with cloud features disabled.","website":"https://ollama.com","repository":"https://github.com/ollama/ollama","environment":"Local Ollama server with OLLAMA_NO_CLOUD=1 and preloaded local models; excludes Ollama Cloud and web search.","contributor":"Testedknowhow model runtime research","affiliation":"Commissioned by a Nimbalyst maintainer. Documentation synthesis; no independent review or product execution.","contentLicense":"Unlicensed prototype; redistribution decision pending","reviewStatus":"unreviewed","lastChecked":"2026-09-15","expiresAt":"2026-10-15","applicability":{"versions":["Repository revision a43fad18b088095de20fbd7a8f0de50824cf5d27; living documentation separately dated."],"environment":"Local Ollama server with OLLAMA_NO_CLOUD=1 and preloaded local models; excludes Ollama Cloud and web search.","excludes":["Runtime verification and performance rankings","Model weights license or quality evaluation","Cloud inference, remote hosts and external tools","Unlisted model architectures, builds and hardware"]},"claims":[{"id":"ollama-local-rt-local","capability":"rt-local","statement":"Runs local models with cloud features disabled.","method":"documented","outcome":"claimed","scope":"Set OLLAMA_NO_CLOUD=1 and restart; prefetch model files separately.","evidenceIds":["ollama-local-faq"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-apple-silicon","capability":"rt-apple-silicon","statement":"Supports Apple GPU acceleration through Metal.","method":"documented","outcome":"claimed","scope":"Supported Apple hardware; actual memory fit unmeasured.","evidenceIds":["ollama-local-gpu"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-cpu","capability":"rt-cpu","statement":"Documents CPU execution and GPU offload modes.","method":"documented","outcome":"claimed","scope":"Backend placement can be inspected using ollama ps.","evidenceIds":["ollama-local-faq"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-nvidia","capability":"rt-nvidia","statement":"Supports NVIDIA GPUs within documented driver and compute requirements.","method":"documented","outcome":"claimed","scope":"Check the hardware compatibility list for the chosen GPU.","evidenceIds":["ollama-local-gpu"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-gguf","capability":"rt-gguf","statement":"Imports local GGUF model files through a Modelfile.","method":"documented","outcome":"claimed","scope":"Supported model architectures; import does not quantize the GGUF file.","evidenceIds":["ollama-local-import"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-quantized","capability":"rt-quantized","statement":"Can load previously quantized GGUF models.","method":"documented","outcome":"claimed","scope":"Prepare quantization before GGUF import; no quality equivalence asserted.","evidenceIds":["ollama-local-import"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-http","capability":"rt-http","statement":"Exposes a local HTTP API.","method":"documented","outcome":"claimed","scope":"Default bind is 127.0.0.1:11434; changing it affects network exposure.","evidenceIds":["ollama-local-faq"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-openai","capability":"rt-openai","statement":"Supports a subset of OpenAI-compatible API fields and endpoints.","method":"documented","outcome":"claimed","scope":"Only the documented endpoint and parameter subset.","evidenceIds":["ollama-local-openai"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-structured","capability":"rt-structured","statement":"Accepts JSON schemas for structured generation.","method":"documented","outcome":"claimed","scope":"Local runtime only; cited documentation excludes Ollama Cloud support.","evidenceIds":["ollama-local-structured"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-streaming","capability":"rt-streaming","statement":"Streams generated output as newline-delimited JSON.","method":"documented","outcome":"claimed","scope":"Native API streaming protocol; separate from compatible API streaming.","evidenceIds":["ollama-local-streaming"],"contradictions":[],"support":"yes"},{"id":"ollama-local-rt-concurrent","capability":"rt-concurrent","statement":"Can process multiple requests to the same model in parallel.","method":"documented","outcome":"claimed","scope":"Requires sufficient memory; parallel request count multiplies context memory and excess work queues.","evidenceIds":["ollama-local-faq"],"contradictions":[],"support":"yes"}],"evidence":[{"id":"ollama-local-license","url":"https://github.com/ollama/ollama/blob/a43fad18b088095de20fbd7a8f0de50824cf5d27/LICENSE","title":"Ollama runtime license","locator":"MIT License","capturedAt":"2026-09-15","kind":"repository","revision":"a43fad18b088095de20fbd7a8f0de50824cf5d27","redistribution":"Link and original paraphrase only; upstream rights retained."},{"id":"ollama-local-faq","url":"https://docs.ollama.com/faq","title":"Ollama FAQ","locator":"Local data; disabling cloud; server binding; concurrent requests","capturedAt":"2026-09-15","kind":"vendor-documentation","revision":null,"redistribution":"Link and original paraphrase only; upstream rights retained."},{"id":"ollama-local-gpu","url":"https://docs.ollama.com/gpu","title":"Ollama hardware support","locator":"Apple Metal; NVIDIA requirements; CPU fallback","capturedAt":"2026-09-15","kind":"vendor-documentation","revision":null,"redistribution":"Link and original paraphrase only; upstream rights retained."},{"id":"ollama-local-import","url":"https://docs.ollama.com/import","title":"Ollama model import","locator":"Importing a GGUF model; quantization limitations","capturedAt":"2026-09-15","kind":"vendor-documentation","revision":null,"redistribution":"Link and original paraphrase only; upstream rights retained."},{"id":"ollama-local-openai","url":"https://docs.ollama.com/api/openai-compatibility","title":"Ollama OpenAI compatibility","locator":"Supported endpoints and supported fields","capturedAt":"2026-09-15","kind":"vendor-documentation","revision":null,"redistribution":"Link and original paraphrase only; upstream rights retained."},{"id":"ollama-local-structured","url":"https://docs.ollama.com/capabilities/structured-outputs","title":"Ollama structured outputs","locator":"JSON schema format; OpenAI compatibility; cloud limitation","capturedAt":"2026-09-15","kind":"vendor-documentation","revision":null,"redistribution":"Link and original paraphrase only; upstream rights retained."},{"id":"ollama-local-streaming","url":"https://docs.ollama.com/api/streaming","title":"Ollama API streaming","locator":"Newline-delimited JSON streaming","capturedAt":"2026-09-15","kind":"vendor-documentation","revision":null,"redistribution":"Link and original paraphrase only; upstream rights retained."}],"gaps":["Which pinned model, quantization, context length and runtime versions fit the selected Mac?","What are measured latency, throughput, peak memory and swap under one and four concurrent requests?"],"invalidationTriggers":["Runtime, bundled engine, backend, model format or license changes.","Documentation changes affecting the cited configuration."],"verificationPlan":["Choose hardware and obtain explicit approval for installations, downloads and execution resources.","Pin runtime and engine versions, model artifact hashes, tokenizer, template, quantization and inference settings.","Run the local-model-mac investigation with synthetic prompts and capture raw measurements; do not record documentation as execution."],"attestations":[],"space":"model-runtimes"}