--- type: "firecrawl-provider" description: "Ollama's public model library at ollama.com: search and browse official and community models, read model pages, tags with download size and context window, per-tag layers with exact byte sizes from the Ollama registry manifest, and GGUF layer metadata (context_length, embedding_length)." use_when: "Ollama's public model library at ollama.com: search and browse official and community models, read model pages, tags with download size and context window, per-tag layers with exact byte sizes from the Ollama registry manifest, and GGUF layer metadata (context_length, embedding_length)." categories: "AI models" capabilities: 7 credits_per_call: 5 --- # Ollama model library on Firecrawl Alexandria Ollama's public model library at ollama.com: search and browse official and community models, read model pages, tags with download size and context window, per-tag layers with exact byte sizes from the Ollama registry manifest, and GGUF layer metadata (context_length, embedding_length). - Categories: AI models - Category index: [AI models category](https://firecrawl.dev/alexandria/agents/categories/ai-models) - Provider key: `ollama-com` - Access: Firecrawl credits - Cost: 5 credits per call ## More - [Human guide](https://firecrawl.dev/app/alexandria/ollama-com) - [OpenAPI spec](https://firecrawl.dev/alexandria/agents/providers/ollama-com/openapi.json) ## Capabilities - [Layer metadata](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/layer_metadata): GGUF metadata of one model layer (`/{model}:{tag}/blobs/{digest}`): architecture, file type, exact context_length, embedding_length (vector dimension), block_count and every other key the site renders, with the tensor table on request. - [Library models](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/library_models): The official Ollama library (`/library`, namespace `library`): every official model in one response, optionally narrowed by a name filter and sorted by popularity or recency. - [Model](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/model): One model's page: description, capability and size badges, pull count, last update, the featured tag table (size, context window, input modality, `latest` alias) and the readme as text with its links. - [Model tags](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/model_tags): Every tag of a model (`/{model}/tags`): tag name, 12-hex manifest digest, rounded download size, context window, input modality, relative age and the `latest` alias marker. - [Namespace models](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/namespace_models): Models published by one community namespace (`/{namespace}`): profile heading, bio, links and model cards. - [Search models](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/search_models): Search ollama.com for models by free text and/or capability (embedding, vision, tools, thinking, cloud), sorted by popularity or recency. Twenty model cards per page with description, capability and size badges, pull count, tag count and last update. - [Tag](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/tag): One model tag (`/{model}:{tag}`): manifest short digest, rounded size, model-layer architecture / parameter count / quantization, runtime params (e.g. num_ctx) when the params layer is short enough to render, every layer with its blob page, and, from registry.ollama.ai, the OCI manifest with full sha256 digests and exact byte sizes. ## 1. Choose this provider when Ollama's public model library at ollama.com: search and browse official and community models, read model pages, tags with download size and context window, per-tag layers with exact byte sizes from the Ollama registry manifest, and GGUF layer metadata (context_length, embedding_length). ## 2. Minimal request Call `POST https://api.firecrawl.dev/v2/scrape` with `{ alexandria: { provider, capability, options } }`. For a batch, send `{ alexandria: [...] }` with up to 10 calls. ```json { "provider": "ollama-com", "capability": "models/layer_metadata", "options": { "layer_digest": "06507c7b4268", "model": "qwen3-embedding", "tag": "0.6b" } } ``` ## 3. Add provider options Use only the options needed for the task: - `include_tensors` (boolean): Return the tensor table (name, type, shape); a few hundred rows. Default false. Example: `false` - `layer_digest` (string, required): 12-hex short digest from tag.layers[].short_digest (a full sha256 is accepted and shortened). Pattern: ^(sha256:)?[0-9a-fA-F]{12}([0-9a-fA-F]{52})?$. Example: `` - `model` (string, required): Model name as on ollama.com: `nomic-embed-text` (official, namespace `library`) or `namespace/name` for a community model (e.g. `aiconjured/embeddinggemma-300M-NVFP4-Q8-GGUF`). No `:tag`. Pattern: ^[A-Za-z0-9._-]+(/[A-Za-z0-9._-]+)?$. Example: `nomic-embed-text` - `tag` (string, required): Tag after the colon, e.g. `latest`, `0.6b`, `137m-v1.5-fp16`. Pattern: ^[A-Za-z0-9._-]+$. Example: `` ## 4. Request through your preferred interface ### JavaScript ```javascript const result = await firecrawl.scrape({ alexandria: { provider: "ollama-com", capability: "models/layer_metadata", options: { layer_digest: "06507c7b4268", model: "qwen3-embedding", tag: "0.6b", }, }, }); ``` ### Python ```python result = firecrawl.scrape_alexandria({ "provider": "ollama-com", "capability": "models/layer_metadata", "options": { "layer_digest": "06507c7b4268", "model": "qwen3-embedding", "tag": "0.6b" } }) ``` ### cURL ```sh curl https://api.firecrawl.dev/v2/scrape \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "alexandria": { "provider": "ollama-com", "capability": "models/layer_metadata", "options": { "layer_digest": "06507c7b4268", "model": "qwen3-embedding", "tag": "0.6b" } } }' ``` ### CLI ```sh firecrawl scrape 'ollama-com/models/layer_metadata' \ --options '{"layer_digest":"06507c7b4268","model":"qwen3-embedding","tag":"0.6b"}' ``` ### MCP Call the FCX MCP retrieve tool with this object: ```json { "provider": "ollama-com", "capability": "models/layer_metadata", "options": { "layer_digest": "06507c7b4268", "model": "qwen3-embedding", "tag": "0.6b" } } ``` Ask for only the returned fields needed by the task. ## 5. Full request shape ```json { "provider": "ollama-com", "capability": "models/layer_metadata", "options": { "layer_digest": "06507c7b4268", "model": "qwen3-embedding", "tag": "0.6b" } } ``` ## 6. Response data The response includes `success`, `provider`, `capability`, `creditsCost` and `data`. This example shows the provider payload in `data`: ```json { "architecture": "qwen3", "block_count": 28, "context_length": 32768, "embedding_length": 1024, "file_type": "Q8_0", "layer_type": "model", "metadata": { "general.architecture": "qwen3", "general.file_type": "Q8_0", "qwen3.attention.head_count": "16", "qwen3.attention.head_count_kv": "8", "qwen3.attention.key_length": "128", "qwen3.attention.layer_norm_rms_epsilon": "1e-06", "qwen3.attention.value_length": "128", "qwen3.block_count": "28", "qwen3.context_length": "32768", "qwen3.embedding_length": "1024", "qwen3.feed_forward_length": "3072", "qwen3.pooling_type": "Last", "qwen3.rope.freq_base": "1e+06", "tokenizer.ggml.add_bos_token": "false", "tokenizer.ggml.add_eos_token": "true", "tokenizer.ggml.bos_token_id": "151643", "tokenizer.ggml.eos_token_id": "151643", "tokenizer.ggml.eot_token_id": "151645", "tokenizer.ggml.merges": "[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]", "tokenizer.ggml.model": "gpt2", "tokenizer.ggml.padding_token_id": "151643", "tokenizer.ggml.pre": "qwen2", "tokenizer.ggml.token_type": "[1, 1, 1, 1, 1, ...]", "tokenizer.ggml.tokens": "[!, \", #, $, %, ...]" }, "model": "qwen3-embedding", "name": "qwen3-embedding:0.6b", "namespace": "library", "observed_at_ms": 1790115776734, "short_digest": "06507c7b4268", "size_text": "639MB", "tag": "0.6b", "tensor_count": 310, "tensors": null, "url": "https://ollama.com/library/qwen3-embedding:0.6b/blobs/06507c7b4268" } ``` ## API reference-derived contract The following capability contract is generated from the same normalized Alexandria API reference exposed in the API spec. ### Layer metadata - Capability: `models/layer_metadata` - Description: GGUF metadata of one model layer (`/{model}:{tag}/blobs/{digest}`): architecture, file type, exact context_length, embedding_length (vector dimension), block_count and every other key the site renders, with the tensor table on request. - Instructions: Compare embedding models on true context length and vector dimension, or inspect quantisation per tensor; the `layer_digest` is a model layer's `short_digest` from tag. - Cost: 5 credits per call - Capability file: [Layer metadata](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/layer_metadata) Accepted options: - `include_tensors` (boolean): Return the tensor table (name, type, shape); a few hundred rows. Default false. Example: `false` - `layer_digest` (string, required): 12-hex short digest from tag.layers[].short_digest (a full sha256 is accepted and shortened). Pattern: ^(sha256:)?[0-9a-fA-F]{12}([0-9a-fA-F]{52})?$. Example: `` - `model` (string, required): Model name as on ollama.com: `nomic-embed-text` (official, namespace `library`) or `namespace/name` for a community model (e.g. `aiconjured/embeddinggemma-300M-NVFP4-Q8-GGUF`). No `:tag`. Pattern: ^[A-Za-z0-9._-]+(/[A-Za-z0-9._-]+)?$. Example: `nomic-embed-text` - `tag` (string, required): Tag after the colon, e.g. `latest`, `0.6b`, `137m-v1.5-fp16`. Pattern: ^[A-Za-z0-9._-]+$. Example: `` Response schema example: ```json { "architecture": "qwen3", "block_count": 28, "context_length": 32768, "embedding_length": 1024, "file_type": "Q8_0", "layer_type": "model", "metadata": { "general.architecture": "qwen3", "general.file_type": "Q8_0", "qwen3.attention.head_count": "16", "qwen3.attention.head_count_kv": "8", "qwen3.attention.key_length": "128", "qwen3.attention.layer_norm_rms_epsilon": "1e-06", "qwen3.attention.value_length": "128", "qwen3.block_count": "28", "qwen3.context_length": "32768", "qwen3.embedding_length": "1024", "qwen3.feed_forward_length": "3072", "qwen3.pooling_type": "Last", "qwen3.rope.freq_base": "1e+06", "tokenizer.ggml.add_bos_token": "false", "tokenizer.ggml.add_eos_token": "true", "tokenizer.ggml.bos_token_id": "151643", "tokenizer.ggml.eos_token_id": "151643", "tokenizer.ggml.eot_token_id": "151645", "tokenizer.ggml.merges": "[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]", "tokenizer.ggml.model": "gpt2", "tokenizer.ggml.padding_token_id": "151643", "tokenizer.ggml.pre": "qwen2", "tokenizer.ggml.token_type": "[1, 1, 1, 1, 1, ...]", "tokenizer.ggml.tokens": "[!, \", #, $, %, ...]" }, "model": "qwen3-embedding", "name": "qwen3-embedding:0.6b", "namespace": "library", "observed_at_ms": 1790115776734, "short_digest": "06507c7b4268", "size_text": "639MB", "tag": "0.6b", "tensor_count": 310, "tensors": null, "url": "https://ollama.com/library/qwen3-embedding:0.6b/blobs/06507c7b4268" } ``` ### Library models - Capability: `models/library_models` - Description: The official Ollama library (`/library`, namespace `library`): every official model in one response, optionally narrowed by a name filter and sorted by popularity or recency. - Instructions: Enumerate official models (about 240) or filter them by a substring such as `embed`; community models are only reachable through search_models or namespace_models. - Cost: 5 credits per call - Capability file: [Library models](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/library_models) Accepted options: - `filter` (string): Name filter applied by the site (`embed`). Example: `` - `sort` (string): Order; the site default is `popular`. Example: `popular` Response schema example: ```json { "count": 8, "filter": "embed", "models": [ { "capabilities": [ "embedding" ], "description": "A high-performing open embedding model with a large token context window.", "model": "nomic-embed-text", "name": "nomic-embed-text", "namespace": "library", "pulls": 86800000, "pulls_text": "86.8M", "sizes": [], "tag_count": 3, "updated_at": "2024-02-21T17:26:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/nomic-embed-text" }, { "capabilities": [ "embedding" ], "description": "State-of-the-art large embedding model from mixedbread.ai", "model": "mxbai-embed-large", "name": "mxbai-embed-large", "namespace": "library", "pulls": 14900000, "pulls_text": "14.9M", "sizes": [ "335m" ], "tag_count": 4, "updated_at": "2024-05-06T23:36:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/mxbai-embed-large" }, { "capabilities": [ "embedding" ], "description": "Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes", "model": "qwen3-embedding", "name": "qwen3-embedding", "namespace": "library", "pulls": 4100000, "pulls_text": "4.1M", "sizes": [ "0.6b", "4b", "8b" ], "tag_count": 12, "updated_at": "2025-09-23T20:26:00Z", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding" }, { "capabilities": [ "embedding" ], "description": "A suite of text embedding models by Snowflake, optimized for performance.", "model": "snowflake-arctic-embed", "name": "snowflake-arctic-embed", "namespace": "library", "pulls": 3100000, "pulls_text": "3.1M", "sizes": [ "22m", "33m", "110m", "137m", "335m" ], "tag_count": 16, "updated_at": "2024-04-16T16:06:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/snowflake-arctic-embed" }, { "capabilities": [ "embedding" ], "description": "EmbeddingGemma is a 300M parameter embedding model from Google.", "model": "embeddinggemma", "name": "embeddinggemma", "namespace": "library", "pulls": 2200000, "pulls_text": "2.2M", "sizes": [ "300m" ], "tag_count": 5, "updated_at": "2025-09-09T00:17:00Z", "updated_relative": "1 year ago", "url": "https://ollama.com/library/embeddinggemma" }, { "capabilities": [ "embedding" ], "description": "nomic-embed-text-v2-moe is a multilingual MoE text embedding model that excels at multilingual retrieval.", "model": "nomic-embed-text-v2-moe", "name": "nomic-embed-text-v2-moe", "namespace": "library", "pulls": 938500, "pulls_text": "938.5K", "sizes": [], "tag_count": null, "updated_at": "2025-12-10T22:09:00Z", "updated_relative": "9 months ago", "url": "https://ollama.com/library/nomic-embed-text-v2-moe" }, { "capabilities": [ "embedding" ], "description": "Snowflake's frontier embedding model. Arctic Embed 2.0 adds multilingual support without sacrificing English performance or scalability.", "model": "snowflake-arctic-embed2", "name": "snowflake-arctic-embed2", "namespace": "library", "pulls": 455800, "pulls_text": "455.8K", "sizes": [ "568m" ], "tag_count": 3, "updated_at": "2024-12-04T23:57:00Z", "updated_relative": "1 year ago", "url": "https://ollama.com/library/snowflake-arctic-embed2" }, { "capabilities": [ "embedding" ], "description": "The IBM Granite Embedding 30M and 278M models models are text-only dense biencoder embedding models, with 30M available in English only and 278M serving multilingual use cases.", "model": "granite-embedding", "name": "granite-embedding", "namespace": "library", "pulls": 372400, "pulls_text": "372.4K", "sizes": [ "30m", "278m" ], "tag_count": 6, "updated_at": "2024-12-18T04:11:00Z", "updated_relative": "1 year ago", "url": "https://ollama.com/library/granite-embedding" } ], "observed_at_ms": 1790115752220, "sort": "popular", "source_url": "https://ollama.com/library?sort=popular&q=embed" } ``` ### Model - Capability: `models/model` - Description: One model's page: description, capability and size badges, pull count, last update, the featured tag table (size, context window, input modality, `latest` alias) and the readme as text with its links. - Instructions: Read a model found by search_models or library_models; use model_tags for every tag and tag for one tag's layers and exact byte sizes. - Cost: 5 credits per call - Capability file: [Model](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/model) Accepted options: - `model` (string, required): Model name as on ollama.com: `nomic-embed-text` (official, namespace `library`) or `namespace/name` for a community model (e.g. `aiconjured/embeddinggemma-300M-NVFP4-Q8-GGUF`). No `:tag`. Pattern: ^[A-Za-z0-9._-]+(/[A-Za-z0-9._-]+)?$. Example: `nomic-embed-text` Response schema example: ```json { "capabilities": [ "embedding" ], "description": "A high-performing open embedding model with a large token context window.", "model": "nomic-embed-text", "name": "nomic-embed-text", "namespace": "library", "observed_at_ms": 1790115761863, "pulls": 86800000, "pulls_text": "86.8M", "readme_links": [ { "label": "Download it here", "url": "https://ollama.com/download" }, { "label": "HuggingFace", "url": "https://huggingface.co/nomic-ai/nomic-embed-text-v1.5" }, { "label": "Blog Post", "url": "https://blog.nomic.ai/posts/nomic-embed-text-v1" } ], "readme_text": "Note: this model requires Ollama 0.1.26 or later. Download it here . It can only be used to generate embeddings. nomic-embed-text is a large context length text encoder that surpasses OpenAI text-embedding-ada-002 and text-embedding-3-small performance on short and long context tasks. Usage This model is an embedding model, meaning it can only be used to generate embeddings. REST API curl http://localhost:11434/api/embeddings -d '{ \"model\": \"nomic-embed-text\", \"prompt\": \"The sky is blue because of Rayleigh scattering\" }' Python library ollama.embeddings(model='nomic-embed-text', prompt='The sky is blue because of rayleigh scattering') Javascript library ollama.embeddings({ model: 'nomic-embed-text', prompt: 'The sky is blue because of rayleigh scattering' }) References HuggingFace Blog Post", "sizes": [], "tags": [ { "context_text": "2K", "input": "Text", "is_latest_alias": false, "name": "nomic-embed-text:latest", "size_text": "274MB", "tag": "latest", "url": "https://ollama.com/library/nomic-embed-text:latest" }, { "context_text": "2K", "input": "Text", "is_latest_alias": true, "name": "nomic-embed-text:v1.5", "size_text": "274MB", "tag": "v1.5", "url": "https://ollama.com/library/nomic-embed-text:v1.5" }, { "context_text": "2K", "input": "Text", "is_latest_alias": false, "name": "nomic-embed-text:137m-v1.5-fp16", "size_text": "274MB", "tag": "137m-v1.5-fp16", "url": "https://ollama.com/library/nomic-embed-text:137m-v1.5-fp16" } ], "tags_url": "https://ollama.com/library/nomic-embed-text/tags", "updated_at": "2024-02-21T17:26:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/nomic-embed-text" } ``` ### Model tags - Capability: `models/model_tags` - Description: Every tag of a model (`/{model}/tags`): tag name, 12-hex manifest digest, rounded download size, context window, input modality, relative age and the `latest` alias marker. - Instructions: Compare a model's variants (quantisations, parameter sizes) or pick the tag to pass to tag / layer_metadata. - Cost: 5 credits per call - Capability file: [Model tags](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/model_tags) Accepted options: - `model` (string, required): Model name as on ollama.com: `nomic-embed-text` (official, namespace `library`) or `namespace/name` for a community model (e.g. `aiconjured/embeddinggemma-300M-NVFP4-Q8-GGUF`). No `:tag`. Pattern: ^[A-Za-z0-9._-]+(/[A-Za-z0-9._-]+)?$. Example: `nomic-embed-text` Response schema example: ```json { "capabilities": [ "embedding" ], "count": 12, "description": "Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes", "model": "qwen3-embedding", "name": "qwen3-embedding", "namespace": "library", "observed_at_ms": 1790115766691, "pulls": 4100000, "pulls_text": "4.1M", "sizes": [ "0.6b", "4b", "8b" ], "tags": [ { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:latest", "short_digest": "64b933495768", "size_text": "4.7GB", "tag": "latest", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:latest" }, { "context_text": "32K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:0.6b", "short_digest": "ac6da0dfba84", "size_text": "639MB", "tag": "0.6b", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:0.6b" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:4b", "short_digest": "df5bd2e3c74c", "size_text": "2.5GB", "tag": "4b", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:4b" }, { "context_text": "40K", "input": "Text", "is_latest_alias": true, "name": "qwen3-embedding:8b", "short_digest": "64b933495768", "size_text": "4.7GB", "tag": "8b", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:8b" }, { "context_text": "32K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:0.6b-q8_0", "short_digest": "ac6da0dfba84", "size_text": "639MB", "tag": "0.6b-q8_0", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:0.6b-q8_0" }, { "context_text": "32K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:0.6b-fp16", "short_digest": "67a7592a8852", "size_text": "1.2GB", "tag": "0.6b-fp16", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:0.6b-fp16" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:4b-q4_K_M", "short_digest": "df5bd2e3c74c", "size_text": "2.5GB", "tag": "4b-q4_K_M", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:4b-q4_K_M" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:4b-q8_0", "short_digest": "357d756ba8e5", "size_text": "4.3GB", "tag": "4b-q8_0", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:4b-q8_0" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:4b-fp16", "short_digest": "3c93b6415795", "size_text": "8.0GB", "tag": "4b-fp16", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:4b-fp16" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:8b-q4_K_M", "short_digest": "64b933495768", "size_text": "4.7GB", "tag": "8b-q4_K_M", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:8b-q4_K_M" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:8b-q8_0", "short_digest": "9704fd987c12", "size_text": "8.0GB", "tag": "8b-q8_0", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:8b-q8_0" }, { "context_text": "40K", "input": "Text", "is_latest_alias": false, "name": "qwen3-embedding:8b-fp16", "short_digest": "aa924958585e", "size_text": "15GB", "tag": "8b-fp16", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:8b-fp16" } ], "tags_url": "https://ollama.com/library/qwen3-embedding/tags", "updated_at": "2025-09-23T20:26:00Z", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding" } ``` ### Namespace models - Capability: `models/namespace_models` - Description: Models published by one community namespace (`/{namespace}`): profile heading, bio, links and model cards. - Instructions: List what a publisher has uploaded; the namespace comes from a search card's `namespace`. - Cost: 5 credits per call - Capability file: [Namespace models](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/namespace_models) Accepted options: - `filter` (string): filter Example: `` - `namespace` (string, required): Publisher handle, e.g. `aiconjured`. Not `library`. Pattern: ^[A-Za-z0-9._-]+$. Example: `` - `sort` (string): Order; the site default is `popular`. Example: `popular` Response schema example: ```json { "bio": "NVFP4 LLM Models", "count": 7, "display_name": "aiconjured", "filter": null, "links": [ { "label": "AIconjured", "url": "https://huggingface.co/AIconjured" } ], "models": [ { "capabilities": [ "vision" ], "description": "Text + Vision Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-NVFP4", "model": "Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-NVFP4", "name": "Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-NVFP4", "namespace": "aiconjured", "pulls": 2100, "pulls_text": "2,100", "sizes": [], "tag_count": null, "updated_at": "2026-09-05T06:39:00Z", "updated_relative": "2 weeks ago", "url": "https://ollama.com/aiconjured/Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-NVFP4" }, { "capabilities": [ "vision" ], "description": "Text + Vision Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-Q8-NVFP4", "model": "Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-Q8-NVFP4", "name": "Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-Q8-NVFP4", "namespace": "aiconjured", "pulls": 1192, "pulls_text": "1,192", "sizes": [], "tag_count": null, "updated_at": "2026-08-24T06:30:00Z", "updated_relative": "4 weeks ago", "url": "https://ollama.com/aiconjured/Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-Q8-NVFP4" }, { "capabilities": [ "vision" ], "description": "Text + Vision Qwen3.8-27B-FableColdFusion-735882-HereticUncensored-NEOCODERMAX-MTP-NVFP4-Q8-Q3", "model": "Qwen3.8-27B-FableColdFusion-735882-HereticUncensored-NEOCODERMAX-MTP-NVFP4-Q8-Q3", "name": "Qwen3.8-27B-FableColdFusion-735882-HereticUncensored-NEOCODERMAX-MTP-NVFP4-Q8-Q3", "namespace": "aiconjured", "pulls": 671, "pulls_text": "671", "sizes": [], "tag_count": null, "updated_at": "2026-09-12T18:15:00Z", "updated_relative": "1 week ago", "url": "https://ollama.com/aiconjured/Qwen3.8-27B-FableColdFusion-735882-HereticUncensored-NEOCODERMAX-MTP-NVFP4-Q8-Q3" }, { "capabilities": [ "vision" ], "description": "Text + Vision All credit: https://huggingface.co/esatapedico/Qwen3.8-27B-NVFP4-MTP-GGUF", "model": "Qwen3.8-27B-NVFP4-MTP-COMPACT-LOW", "name": "Qwen3.8-27B-NVFP4-MTP-COMPACT-LOW", "namespace": "aiconjured", "pulls": 276, "pulls_text": "276", "sizes": [], "tag_count": null, "updated_at": "2026-08-22T03:00:00Z", "updated_relative": "1 month ago", "url": "https://ollama.com/aiconjured/Qwen3.8-27B-NVFP4-MTP-COMPACT-LOW" }, { "capabilities": [ "vision" ], "description": "Text + Vision All credit: https://huggingface.co/esatapedico/Qwen3.8-27B-NVFP4-MTP-GGUF", "model": "Qwen3.8-27B-NVFP4-MTP-VERY-HIGH", "name": "Qwen3.8-27B-NVFP4-MTP-VERY-HIGH", "namespace": "aiconjured", "pulls": 194, "pulls_text": "194", "sizes": [], "tag_count": null, "updated_at": "2026-08-22T03:15:00Z", "updated_relative": "1 month ago", "url": "https://ollama.com/aiconjured/Qwen3.8-27B-NVFP4-MTP-VERY-HIGH" }, { "capabilities": [ "embedding" ], "description": "Embedding Model Qwen3-Embedding-0.6B-Q8-NVFP4", "model": "Qwen3-Embedding-0.6B-Q8-NVFP4", "name": "Qwen3-Embedding-0.6B-Q8-NVFP4", "namespace": "aiconjured", "pulls": 37, "pulls_text": "37", "sizes": [], "tag_count": null, "updated_at": "2026-09-05T07:55:00Z", "updated_relative": "2 weeks ago", "url": "https://ollama.com/aiconjured/Qwen3-Embedding-0.6B-Q8-NVFP4" }, { "capabilities": [ "embedding" ], "description": "Embedding Model embeddinggemma-300M-NVFP4-Q8-GGUF", "model": "embeddinggemma-300M-NVFP4-Q8-GGUF", "name": "embeddinggemma-300M-NVFP4-Q8-GGUF", "namespace": "aiconjured", "pulls": 7, "pulls_text": "7", "sizes": [], "tag_count": null, "updated_at": "2026-09-12T21:28:00Z", "updated_relative": "1 week ago", "url": "https://ollama.com/aiconjured/embeddinggemma-300M-NVFP4-Q8-GGUF" } ], "namespace": "aiconjured", "observed_at_ms": 1790115757027, "sort": "popular", "source_url": "https://ollama.com/aiconjured?sort=popular" } ``` ### Search models - Capability: `models/search_models` - Description: Search ollama.com for models by free text and/or capability (embedding, vision, tools, thinking, cloud), sorted by popularity or recency. Twenty model cards per page with description, capability and size badges, pull count, tag count and last update. - Instructions: Start here to find official and community models; take `namespace` and `model` from a card into model, model_tags and tag. - Cost: 5 credits per call - Capability file: [Search models](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/search_models) Accepted options: - `capability` (string): Restrict to models carrying this capability badge. Example: `cloud` - `page` (number): 1-based page; use `next_page` from the previous result. Example: `10` - `query` (string): Free-text query (name or description words). Example: `` - `sort` (string): Order; the site default is `popular`. Example: `popular` Response schema example: ```json { "capability": null, "count": 20, "has_more": true, "models": [ { "capabilities": [ "embedding" ], "description": "EmbeddingGemma is a 300M parameter embedding model from Google.", "model": "embeddinggemma", "name": "embeddinggemma", "namespace": "library", "pulls": 2200000, "pulls_text": "2.2M", "sizes": [ "300m" ], "tag_count": 5, "updated_at": "2025-09-09T00:17:00Z", "updated_relative": "1 year ago", "url": "https://ollama.com/library/embeddinggemma" }, { "capabilities": [ "embedding" ], "description": "Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes", "model": "qwen3-embedding", "name": "qwen3-embedding", "namespace": "library", "pulls": 4100000, "pulls_text": "4.1M", "sizes": [ "0.6b", "4b", "8b" ], "tag_count": 12, "updated_at": "2025-09-23T20:26:00Z", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding" }, { "capabilities": [ "embedding" ], "description": "nomic-embed-text-v2-moe is a multilingual MoE text embedding model that excels at multilingual retrieval.", "model": "nomic-embed-text-v2-moe", "name": "nomic-embed-text-v2-moe", "namespace": "library", "pulls": 938500, "pulls_text": "938.5K", "sizes": [], "tag_count": null, "updated_at": "2025-12-10T22:09:00Z", "updated_relative": "9 months ago", "url": "https://ollama.com/library/nomic-embed-text-v2-moe" }, { "capabilities": [ "embedding" ], "description": "A high-performing open embedding model with a large token context window.", "model": "nomic-embed-text", "name": "nomic-embed-text", "namespace": "library", "pulls": 86800000, "pulls_text": "86.8M", "sizes": [], "tag_count": 3, "updated_at": "2024-02-21T17:26:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/nomic-embed-text" }, { "capabilities": [ "embedding" ], "description": "State-of-the-art large embedding model from mixedbread.ai", "model": "mxbai-embed-large", "name": "mxbai-embed-large", "namespace": "library", "pulls": 14900000, "pulls_text": "14.9M", "sizes": [ "335m" ], "tag_count": 4, "updated_at": "2024-05-06T23:36:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/mxbai-embed-large" }, { "capabilities": [ "embedding" ], "description": "Embedding models on very large sentence level datasets.", "model": "all-minilm", "name": "all-minilm", "namespace": "library", "pulls": 3600000, "pulls_text": "3.6M", "sizes": [ "22m", "33m" ], "tag_count": 10, "updated_at": "2024-05-06T23:38:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/all-minilm" }, { "capabilities": [ "embedding" ], "description": "A suite of text embedding models by Snowflake, optimized for performance.", "model": "snowflake-arctic-embed", "name": "snowflake-arctic-embed", "namespace": "library", "pulls": 3100000, "pulls_text": "3.1M", "sizes": [ "22m", "33m", "110m", "137m", "335m" ], "tag_count": 16, "updated_at": "2024-04-16T16:06:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/snowflake-arctic-embed" }, { "capabilities": [ "embedding" ], "description": "Snowflake's frontier embedding model. Arctic Embed 2.0 adds multilingual support without sacrificing English performance or scalability.", "model": "snowflake-arctic-embed2", "name": "snowflake-arctic-embed2", "namespace": "library", "pulls": 455800, "pulls_text": "455.8K", "sizes": [ "568m" ], "tag_count": 3, "updated_at": "2024-12-04T23:57:00Z", "updated_relative": "1 year ago", "url": "https://ollama.com/library/snowflake-arctic-embed2" }, { "capabilities": [ "embedding" ], "description": "The IBM Granite Embedding 30M and 278M models models are text-only dense biencoder embedding models, with 30M available in English only and 278M serving multilingual use cases.", "model": "granite-embedding", "name": "granite-embedding", "namespace": "library", "pulls": 372400, "pulls_text": "372.4K", "sizes": [ "30m", "278m" ], "tag_count": 6, "updated_at": "2024-12-18T04:11:00Z", "updated_relative": "1 year ago", "url": "https://ollama.com/library/granite-embedding" }, { "capabilities": [ "embedding" ], "description": "Embedding model from BAAI mapping texts to vectors.", "model": "bge-large", "name": "bge-large", "namespace": "library", "pulls": 286600, "pulls_text": "286.6K", "sizes": [ "335m" ], "tag_count": 3, "updated_at": "2024-08-07T00:04:00Z", "updated_relative": "2 years ago", "url": "https://ollama.com/library/bge-large" }, { "capabilities": [ "tools", "thinking" ], "description": "Laguna XS.2 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.", "model": "laguna-xs.2", "name": "laguna-xs.2", "namespace": "library", "pulls": 30800, "pulls_text": "30.8K", "sizes": [], "tag_count": 7, "updated_at": "2026-07-25T18:19:00Z", "updated_relative": "1 month ago", "url": "https://ollama.com/library/laguna-xs.2" }, { "capabilities": [ "tools", "thinking" ], "description": "An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.", "model": "nemotron-cascade-2", "name": "nemotron-cascade-2", "namespace": "library", "pulls": 148700, "pulls_text": "148.7K", "sizes": [ "30b" ], "tag_count": 3, "updated_at": "2026-03-20T20:10:00Z", "updated_relative": "6 months ago", "url": "https://ollama.com/library/nemotron-cascade-2" }, { "capabilities": [ "embedding" ], "description": "WeMM-Embedding-2B is a universal multimodal embedding model built on Qwen3.5.", "model": "wemm-embedding-2b", "name": "milkey/wemm-embedding-2b", "namespace": "milkey", "pulls": 341, "pulls_text": "341", "sizes": [], "tag_count": 4, "updated_at": "2026-08-26T21:40:00Z", "updated_relative": "3 weeks ago", "url": "https://ollama.com/milkey/wemm-embedding-2b" }, { "capabilities": [ "embedding", "tools" ], "description": null, "model": "Qwen3-Embedding-4B-Q4-K-M", "name": "ttempvnn/Qwen3-Embedding-4B-Q4-K-M", "namespace": "ttempvnn", "pulls": 57, "pulls_text": "57", "sizes": [], "tag_count": null, "updated_at": "2026-09-05T19:19:00Z", "updated_relative": "2 weeks ago", "url": "https://ollama.com/ttempvnn/Qwen3-Embedding-4B-Q4-K-M" }, { "capabilities": [ "embedding", "tools" ], "description": null, "model": "Qwen3-Embedding-8B-Q4-K-M", "name": "ttempvnn/Qwen3-Embedding-8B-Q4-K-M", "namespace": "ttempvnn", "pulls": 54, "pulls_text": "54", "sizes": [], "tag_count": null, "updated_at": "2026-09-03T18:25:00Z", "updated_relative": "2 weeks ago", "url": "https://ollama.com/ttempvnn/Qwen3-Embedding-8B-Q4-K-M" }, { "capabilities": [ "embedding" ], "description": "Embedding Model Qwen3-Embedding-0.6B-Q8-NVFP4", "model": "Qwen3-Embedding-0.6B-Q8-NVFP4", "name": "aiconjured/Qwen3-Embedding-0.6B-Q8-NVFP4", "namespace": "aiconjured", "pulls": 37, "pulls_text": "37", "sizes": [], "tag_count": null, "updated_at": "2026-09-05T07:55:00Z", "updated_relative": "2 weeks ago", "url": "https://ollama.com/aiconjured/Qwen3-Embedding-0.6B-Q8-NVFP4" }, { "capabilities": [ "embedding" ], "description": null, "model": "qwen3-embedding", "name": "batiai/qwen3-embedding", "namespace": "batiai", "pulls": 1890, "pulls_text": "1,890", "sizes": [ "0.6b", "4b", "8b" ], "tag_count": 9, "updated_at": "2026-04-19T23:53:00Z", "updated_relative": "5 months ago", "url": "https://ollama.com/batiai/qwen3-embedding" }, { "capabilities": [ "vision", "embedding", "tools", "thinking" ], "description": null, "model": "qwen3-vl-embedding-8b", "name": "charaf/qwen3-vl-embedding-8b", "namespace": "charaf", "pulls": 1786, "pulls_text": "1,786", "sizes": [], "tag_count": null, "updated_at": "2026-04-04T09:22:00Z", "updated_relative": "5 months ago", "url": "https://ollama.com/charaf/qwen3-vl-embedding-8b" }, { "capabilities": [ "vision", "embedding" ], "description": "Qwen embedding model", "model": "qwen3-vl-embedding-2b", "name": "RizwanMalik/qwen3-vl-embedding-2b", "namespace": "RizwanMalik", "pulls": 364, "pulls_text": "364", "sizes": [], "tag_count": 4, "updated_at": "2026-05-04T10:28:00Z", "updated_relative": "4 months ago", "url": "https://ollama.com/RizwanMalik/qwen3-vl-embedding-2b" }, { "capabilities": [ "tools", "thinking" ], "description": null, "model": "qwen3-embedding-8b-mlx-mxfp8", "name": "charaf/qwen3-embedding-8b-mlx-mxfp8", "namespace": "charaf", "pulls": 215, "pulls_text": "215", "sizes": [], "tag_count": null, "updated_at": "2026-05-10T09:21:00Z", "updated_relative": "4 months ago", "url": "https://ollama.com/charaf/qwen3-embedding-8b-mlx-mxfp8" } ], "next_page": 2, "observed_at_ms": 1790115746739, "page": 1, "query": "embedding", "sort": "popular", "source_url": "https://ollama.com/search?q=embedding&o=" } ``` ### Tag - Capability: `models/tag` - Description: One model tag (`/{model}:{tag}`): manifest short digest, rounded size, model-layer architecture / parameter count / quantization, runtime params (e.g. num_ctx) when the params layer is short enough to render, every layer with its blob page, and, from registry.ollama.ai, the OCI manifest with full sha256 digests and exact byte sizes. - Instructions: Get exact download bytes and layer digests for a tag; pass a model layer's `short_digest` to layer_metadata for GGUF metadata such as context_length and embedding_length. - Cost: 5 credits per call - Capability file: [Tag](https://firecrawl.dev/alexandria/agents/providers/ollama-com/models/tag) Accepted options: - `include_manifest` (boolean): Also fetch the registry manifest (one extra request) for exact byte sizes. Default true. Example: `false` - `model` (string, required): Model name as on ollama.com: `nomic-embed-text` (official, namespace `library`) or `namespace/name` for a community model (e.g. `aiconjured/embeddinggemma-300M-NVFP4-Q8-GGUF`). No `:tag`. Pattern: ^[A-Za-z0-9._-]+(/[A-Za-z0-9._-]+)?$. Example: `nomic-embed-text` - `tag` (string, required): Tag after the colon, e.g. `latest`, `0.6b`, `137m-v1.5-fp16`. Pattern: ^[A-Za-z0-9._-]+$. Example: `` Response schema example: ```json { "architecture": "qwen3", "capabilities": [ "embedding" ], "description": "Building upon the foundational models of the Qwen3 series, Qwen3 Embedding provides a comprehensive range of text embeddings models in various sizes", "layers": [ { "short_digest": "06507c7b4268", "size_text": "639MB", "summary": "arch qwen3 · parameters 596M · quantization Q8_0", "type": "model", "url": "https://ollama.com/library/qwen3-embedding:0.6b/blobs/06507c7b4268" } ], "manifest": { "config": { "digest": "sha256:9202febed9e2dadac14bca089be90864571336fa9f4375b690a26ed548957fde", "media_type": "application/vnd.docker.container.image.v1+json", "size_bytes": 266 }, "layers": [ { "digest": "sha256:06507c7b42688469c4e7298b0a1e16deff06caf291cf0a5b278c308249c3e439", "media_type": "application/vnd.ollama.image.model", "size_bytes": 639150592 } ], "media_type": "application/vnd.docker.distribution.manifest.v2+json", "schema_version": 2, "total_bytes": 639150858, "url": "https://registry.ollama.ai/v2/library/qwen3-embedding/manifests/0.6b" }, "model": "qwen3-embedding", "model_url": "https://ollama.com/library/qwen3-embedding", "name": "qwen3-embedding:0.6b", "namespace": "library", "observed_at_ms": 1790115771802, "parameters": "596M", "params": null, "pulls": 4100000, "pulls_text": "4.1M", "quantization": "Q8_0", "short_digest": "ac6da0dfba84", "size_text": "639MB", "sizes": [ "0.6b", "4b", "8b" ], "tag": "0.6b", "updated_at": "2025-09-23T20:26:00Z", "updated_relative": "12 months ago", "url": "https://ollama.com/library/qwen3-embedding:0.6b" } ```