Ollama: Update APIs, Testcontainers, Documentation

Signed-off-by: Thomas Vitale <ThomasVitale@users.noreply.github.com>

models/spring-ai-ollama/pom.xml

@@ -73,5 +73,11 @@
             <artifactId>junit-jupiter</artifactId>
             <scope>test</scope>
         </dependency>
+
+        <dependency>
+            <groupId>org.testcontainers</groupId>
+            <artifactId>ollama</artifactId>
+            <scope>test</scope>
+        </dependency>
     </dependencies>
 </project>

models/spring-ai-ollama/src/main/java/org/springframework/ai/ollama/api/OllamaApi.java

@@ -139,6 +139,8 @@ public class OllamaApi {
 	 * context will be returned. You may choose to use the raw parameter if you are
 	 * specifying a full templated prompt in your request to the API, and are managing
 	 * history yourself.
+	 * @param images (optional) a list of base64-encoded images (for multimodal models such as llava).
+	 * @param keepAlive (optional) controls how long the model will stay loaded into memory following the request (default: 5m).
 	 */
 	@JsonInclude(Include.NON_NULL)
 	public record GenerateRequest(
@@ -503,9 +505,9 @@ public class OllamaApi {
 	 * @param evalCount number of tokens in the response.
 	 * @param evalDuration time spent generating the response.
 	 * @see <a href=
-	 * "https://github.com/jmorganca/ollama/blob/main/docs/api.md#generate-a-chat-completion">Chat
+	 * "https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-chat-completion">Chat
 	 * Completion API</a>
-	 * @see <a href="https://github.com/jmorganca/ollama/blob/main/api/types.go">Ollama
+	 * @see <a href="https://github.com/ollama/ollama/blob/main/api/types.go">Ollama
 	 * Types</a>
 	 */
 	@JsonInclude(Include.NON_NULL)
@@ -573,6 +575,7 @@ public class OllamaApi {
 	 *
 	 * @param model The name of model to generate embeddings from.
 	 * @param prompt The text to generate embeddings for.
+	 * @param keepAlive Controls how long the model will stay loaded into memory following the request (default: 5m).
 	 * @param options Additional model parameters listed in the documentation for the
 	 * Modelfile such as temperature.
 	 */

models/spring-ai-ollama/src/main/java/org/springframework/ai/ollama/api/OllamaModel.java

@@ -28,6 +28,11 @@ public enum OllamaModel {
 	 */
 	LLAMA2("llama2"),
 
+	/**
+	 * Llama 3 is a collection of language models ranging from 8B and 70B parameters.
+	 */
+	LLAMA3("llama3"),
+
 	/**
 	 * The 7B parameters model
 	 */
@@ -43,6 +48,11 @@ public enum OllamaModel {
 	 */
 	PHI("phi"),
 
+	/**
+	 * The Phi-3 3.8B language model
+	 */
+	PHI3("phi3"),
+
 	/**
 	 * A fine-tuned Mistral model
 	 */

models/spring-ai-ollama/src/main/java/org/springframework/ai/ollama/api/OllamaOptions.java

@@ -35,10 +35,9 @@ import org.springframework.ai.embedding.EmbeddingOptions;
  * @author Christian Tzolov
  * @since 0.8.0
  * @see <a href=
- * "https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">Ollama
+ * "https://github.com/ollama/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">Ollama
  * Valid Parameters and Values</a>
- * @see <a href="https://github.com/jmorganca/ollama/blob/main/api/types.go">Ollama
- * Types</a>
+ * @see <a href="https://github.com/ollama/ollama/blob/main/api/types.go">Ollama Types</a>
  */
 @JsonInclude(Include.NON_NULL)
 public class OllamaOptions implements ChatOptions, EmbeddingOptions {
@@ -47,6 +46,8 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 
 	private static final List<String> NON_SUPPORTED_FIELDS = List.of("model", "format", "keep_alive");
 
+	// Following fields are ptions which must be set when the model is loaded into memory.
+
 	// @formatter:off
 	/**
 	 * useNUMA Whether to use NUMA.
@@ -110,16 +111,6 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 	 */
 	@JsonProperty("use_mlock") private Boolean useMLock;
 
-	/**
-	 * ???
-	 */
-	@JsonProperty("rope_frequency_base") private Float ropeFrequencyBase;
-
-	/**
-	 * ???
-	 */
-	@JsonProperty("rope_frequency_scale") private Float ropeFrequencyScale;
-
 	/**
 	 * Sets the number of threads to use during computation. By default,
 	 * Ollama will detect this for optimal performance. It is recommended to set this
@@ -128,6 +119,8 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 	 */
 	@JsonProperty("num_thread") private Integer numThread;
 
+	// Following fields are predict options used at runtime.
+
 	/**
 	 * ???
 	 */
@@ -156,8 +149,8 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 	/**
 	 * 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)
-		*/
+	 * conservative text. (Default: 0.9)
+	 */
 	@JsonProperty("top_p") private Float topP;
 
 	/**
@@ -208,16 +201,15 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 	@JsonProperty("mirostat") private Integer mirostat;
 
 	/**
-	 * Influences how quickly the algorithm responds to feedback from
-	 * the generated text. A lower learning rate will result in slower adjustments, while
-	 * a higher learning rate will make the algorithm more responsive. (Default: 0.1).
+	 * Controls the balance between coherence and diversity of the output.
+	 * A lower value will result in more focused and coherent text. (Default: 5.0)
 	 */
 	@JsonProperty("mirostat_tau") private Float mirostatTau;
 
 	/**
-	 * Controls the balance between coherence and diversity of the
-	 * output. A lower value will result in more focused and coherent text. (Default:
-	 * 5.0).
+	 * Influences how quickly the algorithm responds to feedback from the generated text.
+	 * A lower learning rate will result in slower adjustments, while a higher learning rate
+	 * will make the algorithm more responsive. (Default: 0.1)
 	 */
 	@JsonProperty("mirostat_eta") private Float mirostatEta;
 
@@ -235,6 +227,7 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 
 
 	// Following fields are not part of the Ollama Options API but part of the Request.
+
 	/**
 	 * NOTE: Synthetic field not part of the official Ollama API.
 	 * Used to allow overriding the model name with prompt options.
@@ -341,16 +334,6 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 		return this;
 	}
 
-	public OllamaOptions withRopeFrequencyBase(Float ropeFrequencyBase) {
-		this.ropeFrequencyBase = ropeFrequencyBase;
-		return this;
-	}
-
-	public OllamaOptions withRopeFrequencyScale(Float ropeFrequencyScale) {
-		this.ropeFrequencyScale = ropeFrequencyScale;
-		return this;
-	}
-
 	public OllamaOptions withNumThread(Integer numThread) {
 		this.numThread = numThread;
 		return this;
@@ -553,22 +536,6 @@ public class OllamaOptions implements ChatOptions, EmbeddingOptions {
 		this.useMLock = useMLock;
 	}
 
-	public Float getRopeFrequencyBase() {
-		return this.ropeFrequencyBase;
-	}
-
-	public void setRopeFrequencyBase(Float ropeFrequencyBase) {
-		this.ropeFrequencyBase = ropeFrequencyBase;
-	}
-
-	public Float getRopeFrequencyScale() {
-		return this.ropeFrequencyScale;
-	}
-
-	public void setRopeFrequencyScale(Float ropeFrequencyScale) {
-		this.ropeFrequencyScale = ropeFrequencyScale;
-	}
-
 	public Integer getNumThread() {
 		return this.numThread;
 	}

models/spring-ai-ollama/src/test/java/org/springframework/ai/ollama/OllamaChatClientIT.java

@@ -29,7 +29,6 @@ import org.junit.jupiter.api.Test;
 import org.springframework.ai.chat.metadata.Usage;
 import org.springframework.ai.chat.prompt.ChatOptionsBuilder;
 import org.springframework.ai.chat.messages.AssistantMessage;
-import org.testcontainers.containers.GenericContainer;
 import org.testcontainers.junit.jupiter.Container;
 import org.testcontainers.junit.jupiter.Testcontainers;
 
@@ -50,6 +49,7 @@ import org.springframework.boot.SpringBootConfiguration;
 import org.springframework.boot.test.context.SpringBootTest;
 import org.springframework.context.annotation.Bean;
 import org.springframework.core.convert.support.DefaultConversionService;
+import org.testcontainers.ollama.OllamaContainer;
 
 import static org.assertj.core.api.Assertions.assertThat;
 
@@ -63,7 +63,7 @@ class OllamaChatClientIT {
 	private static final Log logger = LogFactory.getLog(OllamaChatClientIT.class);
 
 	@Container
-	static GenericContainer<?> ollamaContainer = new GenericContainer<>("ollama/ollama:0.1.29").withExposedPorts(11434);
+	static OllamaContainer ollamaContainer = new OllamaContainer("ollama/ollama:0.1.32");
 
 	static String baseUrl;
 

models/spring-ai-ollama/src/test/java/org/springframework/ai/ollama/OllamaChatClientMultimodalIT.java

@@ -23,7 +23,6 @@ import org.apache.commons.logging.LogFactory;
 import org.junit.jupiter.api.BeforeAll;
 import org.junit.jupiter.api.Disabled;
 import org.junit.jupiter.api.Test;
-import org.testcontainers.containers.GenericContainer;
 import org.testcontainers.junit.jupiter.Container;
 import org.testcontainers.junit.jupiter.Testcontainers;
 
@@ -39,6 +38,7 @@ import org.springframework.boot.test.context.SpringBootTest;
 import org.springframework.context.annotation.Bean;
 import org.springframework.core.io.ClassPathResource;
 import org.springframework.util.MimeTypeUtils;
+import org.testcontainers.ollama.OllamaContainer;
 
 import static org.assertj.core.api.Assertions.assertThat;
 
@@ -52,7 +52,7 @@ class OllamaChatClientMultimodalIT {
 	private static final Log logger = LogFactory.getLog(OllamaChatClientIT.class);
 
 	@Container
-	static GenericContainer<?> ollamaContainer = new GenericContainer<>("ollama/ollama:0.1.29").withExposedPorts(11434);
+	static OllamaContainer ollamaContainer = new OllamaContainer("ollama/ollama:0.1.32");
 
 	static String baseUrl;
 

models/spring-ai-ollama/src/test/java/org/springframework/ai/ollama/OllamaEmbeddingClientIT.java

@@ -34,6 +34,7 @@ import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.boot.SpringBootConfiguration;
 import org.springframework.boot.test.context.SpringBootTest;
 import org.springframework.context.annotation.Bean;
+import org.testcontainers.ollama.OllamaContainer;
 
 import static org.assertj.core.api.Assertions.assertThat;
 
@@ -45,7 +46,7 @@ class OllamaEmbeddingClientIT {
 	private static final Log logger = LogFactory.getLog(OllamaApiIT.class);
 
 	@Container
-	static GenericContainer<?> ollamaContainer = new GenericContainer<>("ollama/ollama:0.1.29").withExposedPorts(11434);
+	static OllamaContainer ollamaContainer = new OllamaContainer("ollama/ollama:0.1.32");
 
 	static String baseUrl;
 

models/spring-ai-ollama/src/test/java/org/springframework/ai/ollama/api/OllamaApiIT.java

@@ -24,9 +24,9 @@ import org.apache.commons.logging.LogFactory;
 import org.junit.jupiter.api.BeforeAll;
 import org.junit.jupiter.api.Disabled;
 import org.junit.jupiter.api.Test;
-import org.testcontainers.containers.GenericContainer;
 import org.testcontainers.junit.jupiter.Container;
 import org.testcontainers.junit.jupiter.Testcontainers;
+import org.testcontainers.ollama.OllamaContainer;
 import reactor.core.publisher.Flux;
 
 import org.springframework.ai.ollama.api.OllamaApi.ChatRequest;
@@ -50,7 +50,7 @@ public class OllamaApiIT {
 	private static final Log logger = LogFactory.getLog(OllamaApiIT.class);
 
 	@Container
-	static GenericContainer<?> ollamaContainer = new GenericContainer<>("ollama/ollama:0.1.29").withExposedPorts(11434);
+	static OllamaContainer ollamaContainer = new OllamaContainer("ollama/ollama:0.1.32");
 
 	static OllamaApi ollamaApi;