@@ -45,14 +45,18 @@ Generate a response for a given prompt with a provided model. This is a streamin
Advanced parameters (optional):
-`format`: the format to return a response in. Currently the only accepted value is `json`
-`format`: the format to return a response in. Format can be `json` or a JSON schema
-`options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`
-`system`: system message to (overrides what is defined in the `Modelfile`)
-`template`: the prompt template to use (overrides what is defined in the `Modelfile`)
-`context`: the context parameter returned from a previous request to `/generate`, this can be used to keep a short conversational memory
-`stream`: if `false` the response will be returned as a single response object, rather than a stream of objects
-`raw`: if `true` no formatting will be applied to the prompt. You may choose to use the `raw` parameter if you are specifying a full templated prompt in your request to the API
-`keep_alive`: controls how long the model will stay loaded into memory following the request (default: `5m`)
-`context` (deprecated): the context parameter returned from a previous request to `/generate`, this can be used to keep a short conversational memory
#### Structured outputs
Structured outputs are supported by providing a JSON schema in the `format` parameter. The model will generate a response that matches the schema. See the [structured outputs](#request-structured-outputs) example below.
@@ -456,11 +506,15 @@ The `message` object has the following fields:
Advanced parameters (optional):
-`format`: the format to return a response in. Currently the only accepted value is `json`
-`format`: the format to return a response in. Format can be `json` or a JSON schema.
-`options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`
-`stream`: if `false` the response will be returned as a single response object, rather than a stream of objects
-`keep_alive`: controls how long the model will stay loaded into memory following the request (default: `5m`)
### Structured outputs
Structured outputs are supported by providing a JSON schema in the `format` parameter. The model will generate a response that matches the schema. See the [Chat request (Structured outputs)](#chat-request-structured-outputs) example below.
Send a chat message with a conversation history. You can use this same approach to start the conversation using multi-shot or chain-of-thought prompting.
@@ -151,7 +151,7 @@ Refer to the section [above](#how-do-i-configure-ollama-server) for how to set e
Ollama runs an HTTP server and can be exposed using a proxy server such as Nginx. To do so, configure the proxy to forward requests and optionally set required headers (if not exposing Ollama on the network). For example, with Nginx:
```
```nginx
server{
listen80;
server_nameexample.com;# Replace with your domain or IP
...
...
@@ -285,4 +285,28 @@ Note: Windows with Radeon GPUs currently default to 1 model maximum due to limit
## How does Ollama load models on multiple GPUs?
Installing multiple GPUs of the same brand can be a great way to increase your available VRAM to load larger models. When you load a new model, Ollama evaluates the required VRAM for the model against what is currently available. If the model will entirely fit on any single GPU, Ollama will load the model on that GPU. This typically provides the best performance as it reduces the amount of data transfering across the PCI bus during inference. If the model does not fit entirely on one GPU, then it will be spread across all the available GPUs.
When loading a new model, Ollama evaluates the required VRAM for the model against what is currently available. If the model will entirely fit on any single GPU, Ollama will load the model on that GPU. This typically provides the best performance as it reduces the amount of data transferring across the PCI bus during inference. If the model does not fit entirely on one GPU, then it will be spread across all the available GPUs.
## How can I enable Flash Attention?
Flash Attention is a feature of most modern models that can significantly reduce memory usage as the context size grows. To enable Flash Attention, set the `OLLAMA_FLASH_ATTENTION` environment variable to `1` when starting the Ollama server.
## How can I set the quantization type for the K/V cache?
The K/V context cache can be quantized to significantly reduce memory usage when Flash Attention is enabled.
To use quantized K/V cache with Ollama you can set the following environment variable:
-`OLLAMA_KV_CACHE_TYPE` - The quantization type for the K/V cache. Default is `f16`.
> Note: Currently this is a global option - meaning all models will run with the specified quantization type.
The currently available K/V cache quantization types are:
-`f16` - high precision and memory usage (default).
-`q8_0` - 8-bit quantization, uses approximately 1/2 the memory of `f16` with a very small loss in precision, this usually has no noticeable impact on the model's quality (recommended if not using f16).
-`q4_0` - 4-bit quantization, uses approximately 1/4 the memory of `f16` with a small-medium loss in precision that may be more noticeable at higher context sizes.
How much the cache quantization impacts the model's response quality will depend on the model and the task. Models that have a high GQA count (e.g. Qwen2) may see a larger impact on precision from quantization than models with a low GQA count.
You may need to experiment with different quantization types to find the best balance between memory usage and quality.
| seed | Sets the random number seed to use for generation. Setting this to a specific number will make the model generate the same text for the same prompt. (Default: 0) | int | seed 42 |
| stop | Sets the stop sequences to use. When this pattern is encountered the LLM will stop generating text and return. Multiple stop patterns may be set by specifying multiple separate `stop` parameters in a modelfile. | string | stop "AI assistant:" |
| tfs_z | Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting. (default: 1) | float | tfs_z 1 |
| num_predict | Maximum number of tokens to predict when generating text. (Default: 128, -1 = infinite generation, -2 = fill context) | int | num_predict 42 |
| num_predict | Maximum number of tokens to predict when generating text. (Default: -1, infinite generation) | int | num_predict 42 |
| top_k | Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative. (Default: 40) | int | top_k 40 |
| top_p | Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9) | float | top_p 0.9 |
| min_p | Alternative to the top_p, and aims to ensure a balance of quality and variety. The parameter *p* represents the minimum probability for a token to be considered, relative to the probability of the most likely token. For example, with *p*=0.05 and the most likely token having a probability of 0.9, logits with a value less than 0.045 are filtered out. (Default: 0.0) | float | min_p 0.05 |
{"role":"user","content":"I have two friends. The first is Ollama 22 years old busy saving the world, and the second is Alonso 23 years old and wants to hang out. Return a list of friends in JSON format"}
