mirror of
https://github.com/infiniflow/ragflow.git
synced 2026-05-25 10:26:59 +08:00
4b96362092
### What problem does this PR solve? The NVIDIA Go driver in `internal/entity/models/nvidia.go` shipped with a stub `Encode` method that returned `no such method`. `conf/models/nvidia.json` already lists `nvidia/llama-3.2-nemoretriever-1b-vlm-embed-v1` as an embedding model, but the conf had no `embedding` URL suffix, so the picker had nothing wired even if `Encode` worked. A tenant who wanted to use NVIDIA NIM for chat (already working) and embeddings from a single provider could not, even though the upstream endpoint is public at `https://integrate.api.nvidia.com/v1/embeddings` and uses an OpenAI-compatible request body extended with the NVIDIA-specific `input_type` and `truncate` fields. Several other Go drivers already implement `Encode` (siliconflow, zhipu-ai, aliyun), so the interface and the pattern are well-established. This PR fills the gap. ### What this PR includes * `conf/models/nvidia.json`: declare the `embedding` URL suffix alongside the existing `chat` and `models` entries. The embedding model entry was already present, so no model addition is needed. * `internal/entity/models/nvidia.go`: replace the `Encode` stub with a real implementation. Adds a small local response type that matches the OpenAI-compatible shape NVIDIA NIM returns. No factory change. No interface change. ### How the driver works * Validates `apiConfig` and the API key, validates the model name, resolves the region with a default fallback (matching the pattern the merged `ListModels` and `CheckConnection` paths in this driver already use), and builds the URL from `BaseURL[region] + URLSuffix.Embedding`. * Sends all input texts in one request as the `input` array, with the NVIDIA-specific `input_type: "query"`, `encoding_format: "float"`, and `truncate: "END"` fields, mirroring the Python `NvidiaEmbed` reference. * Parses `data[*].embedding` and copies each slice into `[][]float64` indexed by `data[*].index` so the output order matches the input order even if the API returns items in a different order. * Handles both `float64` and `float32` element types. * Empty input returns `[][]float64{}` with no HTTP call. * Non-200 responses propagate the upstream status line and body. * A final pass checks every input slot got a vector and returns a clear error if any slot is still nil. * Per-call 30s context deadline so a slow call cannot block forever. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### How was this tested? * `go build ./internal/entity/models/...` returns exit 0. * `go vet ./internal/entity/models/...` is clean. * `gofmt -l internal/entity/models/nvidia.go` is clean. * The full method set on `NvidiaModel` still matches the `ModelDriver` interface. * Pattern parity with the just-merged Aliyun `Encode` (#14647). Closes #14699
504 lines
9.6 KiB
JSON
504 lines
9.6 KiB
JSON
{
|
|
"name": "Nvidia",
|
|
"url": {
|
|
"default": "https://integrate.api.nvidia.com/v1"
|
|
},
|
|
"url_suffix": {
|
|
"chat": "chat/completions",
|
|
"models": "models",
|
|
"embedding": "embeddings"
|
|
},
|
|
"class": "nvidia",
|
|
"models": [
|
|
{
|
|
"name": "abacusai/dracarys-llama-3.1-70b-instruct",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "baai/bge-m3",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "bytedance/seed-oss-36b-instruct",
|
|
"max_tokens": 32768,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "deepseek-ai/deepseek-v4-flash",
|
|
"max_tokens": 1048576,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "deepseek-ai/deepseek-v4-pro",
|
|
"max_tokens": 1048576,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "deepseek-ai/deepseek-v3.2",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "deepseek-ai/deepseek-v3.1",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "google/codegemma-7b",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "google/gemma-2-2b-it",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "google/gemma-4-31b-it",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "google/gemma-7b",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "ibm/granite-3.3-8b-instruct",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "meta/llama-3.1-405b-instruct",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "meta/llama-3.2-90b-vision-instruct",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat",
|
|
"vision"
|
|
]
|
|
},
|
|
{
|
|
"name": "meta/llama-4-maverick-17b-128e-instruct",
|
|
"max_tokens": 1048576,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "microsoft/phi-4-mini-flash-reasoning",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "minimaxai/minimax-m2.1",
|
|
"max_tokens": 204800,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "minimaxai/minimax-m2.5",
|
|
"max_tokens": 204800,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "minimaxai/minimax-m2.7",
|
|
"max_tokens": 204800,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/devstral-2-123b-instruct-2512",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/magistral-small-2506",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/mistral-7b-instruct-v0.3",
|
|
"max_tokens": 32768,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/mistral-large-3-675b-instruct-2512",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/mistral-medium-3-5-128b",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat",
|
|
"vision"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/mistral-nemotron",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "mistralai/mixtral-8x22b-instruct",
|
|
"max_tokens": 65536,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "moonshotai/kimi-k2.5",
|
|
"max_tokens": 262144,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "moonshotai/kimi-k2.6",
|
|
"max_tokens": 262144,
|
|
"model_types": [
|
|
"chat",
|
|
"vision"
|
|
]
|
|
},
|
|
{
|
|
"name": "moonshotai/kimi-k2-instruct",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "moonshotai/kimi-k2-instruct-0905",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "moonshotai/kimi-k2-thinking",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "nvidia/gliner-pii",
|
|
"max_tokens": 4096,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.1-nemoguard-8b-content-safety",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.1-nemoguard-8b-topic-control",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.1-nemotron-nano-8b-v1",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.1-nemotron-safety-guard-8b-v3",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.1-nemotron-ultra-253b-v1",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.2-nemoretriever-1b-vlm-embed-v1",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.2-nv-embedqa-1b-v2",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.3-nemotron-super-49b-v1",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/llama-3.3-nemotron-super-49b-v1.5",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "nvidia/nemoguard-jailbreak-detect",
|
|
"max_tokens": 4096,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nemotron-3-nano-30b-a3b",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat",
|
|
"vision"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "nvidia/nemotron-3-super-120b-a12b",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nemotron-content-safety-reasoning-4b",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nemotron-mini-4b-instruct",
|
|
"max_tokens": 4096,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nv-embed-v1",
|
|
"max_tokens": 32768,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nv-embedqa-e5-v5",
|
|
"max_tokens": 512,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nv-embedqa-mistral-7b-v2",
|
|
"max_tokens": 512,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/nvidia-nemotron-nano-9b-v2",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/riva-translate-4b-instruct-v1_1",
|
|
"max_tokens": 4096,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "nvidia/usdcode",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "openai/gpt-oss-120b",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "qwen/qwen2.5-coder-7b-instruct",
|
|
"max_tokens": 32768,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "qwen/qwen3-5-122b-a10b",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "qwen/qwen3-235b-a22b",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "qwen/qwen3-coder-480b-a35b-instruct",
|
|
"max_tokens": 262144,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "snowflake/arctic-embed-l",
|
|
"max_tokens": 512,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "z-ai/glm-5",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "z-ai/glm-5.1",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
},
|
|
{
|
|
"name": "z-ai/glm-4.7",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
],
|
|
"thinking": {
|
|
"default_value": true,
|
|
"clear_thinking": true
|
|
}
|
|
}
|
|
]
|
|
} |