diff --git a/packages/fdr-sdk/src/__test__/output/cohere/node.json b/packages/fdr-sdk/src/__test__/output/cohere/node.json index 8e9b0e3ddf..7b1944f755 100644 --- a/packages/fdr-sdk/src/__test__/output/cohere/node.json +++ b/packages/fdr-sdk/src/__test__/output/cohere/node.json @@ -2573,7 +2573,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/v2-api-release", "date": "2024-09-26T00:00:00.000Z", "id": "v2.changelog.2024-9-26-1", "pageId": "pages/changelog/2024-09-26-refresh-models-on-azure.mdx", @@ -2809,7 +2808,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/command-model-nightly-available", "date": "2023-01-17T00:00:00.000Z", "id": "v2.changelog.2023-1-17-1", "pageId": "pages/changelog/2023-01-17-command-r-is-a-scalable-llm-for-business.mdx", @@ -2885,7 +2883,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/coclassify-powered-by-our-representational-model-embeddings", "date": "2022-11-03T00:00:00.000Z", "id": "v2.changelog.2022-11-3-1", "pageId": "pages/changelog/2022-11-03-new-logit-bias-experimental-parameter.mdx", @@ -2979,7 +2976,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/new-extremely-large-model", "date": "2022-04-25T00:00:00.000Z", "id": "v2.changelog.2022-4-25-1", "pageId": "pages/changelog/2022-04-25-updated-small-medium-and-large-generation-models.mdx", @@ -3005,7 +3001,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/classification-endpoint", "date": "2022-03-08T00:00:00.000Z", "id": "v2.changelog.2022-3-8-1", "pageId": "pages/changelog/2022-03-08-finetuning-available-policy-updates.mdx", @@ -3014,7 +3009,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/classification-endpoint", "date": "2022-03-08T00:00:00.000Z", "id": "v2.changelog.2022-3-8-2", "pageId": "pages/changelog/2022-03-08-new-improved-generation-models.mdx", diff --git a/packages/fdr-sdk/src/__test__/output/cohere/slugs-sitemap.json b/packages/fdr-sdk/src/__test__/output/cohere/slugs-sitemap.json index 31e82a51d1..967852f44d 100644 --- a/packages/fdr-sdk/src/__test__/output/cohere/slugs-sitemap.json +++ b/packages/fdr-sdk/src/__test__/output/cohere/slugs-sitemap.json @@ -158,6 +158,7 @@ "v2/changelog/embed-v3-is-multimodal", "v2/changelog/commandr-082024-ft", "v2/changelog/v2-api-release", + "v2/changelog/refresh-models-on-azure", "v2/changelog/command-gets-refreshed", "v2/changelog/force-json-object-response-format", "v2/changelog/release-notes-for-june-10th-2024", @@ -174,17 +175,22 @@ "v2/changelog/model-names-are-changing", "v2/changelog/multilingual-support-for-coclassify", "v2/changelog/command-model-nightly-available", + "v2/changelog/command-r-is-a-scalable-llm-for-business", "v2/changelog/multilingual-text-understanding-model-language-detection", "v2/changelog/model-sizing-update-improvements", "v2/changelog/improvements-to-current-models-new-beta-model-command", "v2/changelog/new-look-for-docs", "v2/changelog/coclassify-powered-by-our-representational-model-embeddings", + "v2/changelog/new-logit-bias-experimental-parameter", "v2/changelog/pricing-update-and-new-dashboard-ui", "v2/changelog/introducing-moderate-beta", "v2/changelog/model-parameter-now-optional", "v2/changelog/new-improved-generation-and-representation-models", "v2/changelog/new-extremely-large-model", + "v2/changelog/updated-small-medium-and-large-generation-models", "v2/changelog/classification-endpoint", + "v2/changelog/finetuning-available-policy-updates", + "v2/changelog/new-improved-generation-models", "v2/changelog/extremely-large-beta-release", "v2/changelog/larger-representation-models", "v2/page/cookbooks", @@ -242,11 +248,5 @@ "v1/docs/faster-web-search", "v1/docs/migrating-from-cogenerate-to-cochat", "v1/docs/classify-starting-the-training", - "v1/changelog", - "v2/changelog/refresh-models-on-azure", - "v2/changelog/command-r-is-a-scalable-llm-for-business", - "v2/changelog/new-logit-bias-experimental-parameter", - "v2/changelog/updated-small-medium-and-large-generation-models", - "v2/changelog/finetuning-available-policy-updates", - "v2/changelog/new-improved-generation-models" + "v1/changelog" ] \ No newline at end of file diff --git a/packages/fdr-sdk/src/__test__/output/cohere/versionNodes.json b/packages/fdr-sdk/src/__test__/output/cohere/versionNodes.json index c7169a4285..2c4a6e4e87 100644 --- a/packages/fdr-sdk/src/__test__/output/cohere/versionNodes.json +++ b/packages/fdr-sdk/src/__test__/output/cohere/versionNodes.json @@ -2571,7 +2571,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/v2-api-release", "date": "2024-09-26T00:00:00.000Z", "id": "v2.changelog.2024-9-26-1", "pageId": "pages/changelog/2024-09-26-refresh-models-on-azure.mdx", @@ -2807,7 +2806,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/command-model-nightly-available", "date": "2023-01-17T00:00:00.000Z", "id": "v2.changelog.2023-1-17-1", "pageId": "pages/changelog/2023-01-17-command-r-is-a-scalable-llm-for-business.mdx", @@ -2883,7 +2881,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/coclassify-powered-by-our-representational-model-embeddings", "date": "2022-11-03T00:00:00.000Z", "id": "v2.changelog.2022-11-3-1", "pageId": "pages/changelog/2022-11-03-new-logit-bias-experimental-parameter.mdx", @@ -2977,7 +2974,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/new-extremely-large-model", "date": "2022-04-25T00:00:00.000Z", "id": "v2.changelog.2022-4-25-1", "pageId": "pages/changelog/2022-04-25-updated-small-medium-and-large-generation-models.mdx", @@ -3003,7 +2999,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/classification-endpoint", "date": "2022-03-08T00:00:00.000Z", "id": "v2.changelog.2022-3-8-1", "pageId": "pages/changelog/2022-03-08-finetuning-available-policy-updates.mdx", @@ -3012,7 +3007,6 @@ "type": "changelogEntry" }, { - "canonicalSlug": "v2/changelog/classification-endpoint", "date": "2022-03-08T00:00:00.000Z", "id": "v2.changelog.2022-3-8-2", "pageId": "pages/changelog/2022-03-08-new-improved-generation-models.mdx", diff --git a/packages/fdr-sdk/src/navigation/migrators/v1ToV2.ts b/packages/fdr-sdk/src/navigation/migrators/v1ToV2.ts index 06297778f0..d15c0caabf 100644 --- a/packages/fdr-sdk/src/navigation/migrators/v1ToV2.ts +++ b/packages/fdr-sdk/src/navigation/migrators/v1ToV2.ts @@ -410,13 +410,7 @@ export class FernNavigationV1ToLatest { ): FernNavigation.ChangelogNode => { const slug = FernNavigation.Slug(node.slug); const overviewPageId = node.overviewPageId ? FernNavigation.PageId(node.overviewPageId) : undefined; - const canonicalSlug = - overviewPageId != null - ? this.#getAndSetCanonicalSlug( - [overviewPageId, this.#createTitleDisambiguationKey(node, parents)], - slug, - ) - : undefined; + const canonicalSlug = overviewPageId != null ? this.#getAndSetCanonicalSlug(overviewPageId, slug) : undefined; const latest: FernNavigation.ChangelogNode = { type: "changelog", id: FernNavigation.NodeId(node.id), @@ -482,10 +476,8 @@ export class FernNavigationV1ToLatest { parents: FernNavigation.V1.NavigationNode[], ): FernNavigation.ChangelogEntryNode => { const slug = FernNavigation.Slug(node.slug); - const canonicalSlug = this.#getAndSetCanonicalSlug( - [node.pageId, this.#createTitleDisambiguationKey(node, parents)], - slug, - ); + // NOTE: do NOT use title disambiguation key here, since the title may not always be unique + const canonicalSlug = this.#getAndSetCanonicalSlug(node.pageId, slug); const latest: FernNavigation.ChangelogEntryNode = { type: "changelogEntry", id: FernNavigation.NodeId(node.id), diff --git a/packages/ui/fern-docs-search-server/src/algolia/__test__/__snapshots__/cohere.test.ts.snap b/packages/ui/fern-docs-search-server/src/algolia/__test__/__snapshots__/cohere.test.ts.snap index 9be6ad3f65..a5324c04e6 100644 --- a/packages/ui/fern-docs-search-server/src/algolia/__test__/__snapshots__/cohere.test.ts.snap +++ b/packages/ui/fern-docs-search-server/src/algolia/__test__/__snapshots__/cohere.test.ts.snap @@ -63465,6 +63465,53 @@ These APIs are in Beta and are subject to updates. We welcome feedback in our Di "slug": "", }, }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2024", + "title": "2024", + }, + { + "slug": "changelog/9", + "title": "September 1900", + }, + ], + "description": "Introducing our improved Command models are available on the Azure cloud computing platform. + +You'll recall that we released refreshed models of Command R and Command R+ in August. +Today, we're pleased to announce that these models are available on the Azure cloud computing platform! +You can find more information about using Cohere's Command models on Azure here.", + "indexSegmentId": "0", + "slug": "v2/changelog/refresh-models-on-azure", + "title": "The refreshed Command R and Command R+ models are now on Azure", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2024", + "September 1900", + ], + "content": "You'll recall that we released refreshed models of Command R and Command R+ in August. +Today, we're pleased to announce that these models are available on the Azure cloud computing platform! +You can find more information about using Cohere's Command models on Azure here.", + "indexSegmentId": "0", + "slug": "v2/changelog/refresh-models-on-azure", + "title": "The refreshed Command R and Command R+ models are now on Azure", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, { "breadcrumbs": [ { @@ -64884,6 +64931,97 @@ If you were previously using the command-xlarge-20221108 model, you will now be "slug": "", }, }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2023", + "title": "2023", + }, + { + "slug": "changelog/1", + "title": "January 1900", + }, + ], + "description": "Explore Command R+, Cohere's powerful language model, excelling in multi-step tool use and complex conversational AI tasks. + +We're pleased to announce the release of Command R+, our newest and most performant large language model. Command R+ is optimized for conversational interaction and long-context tasks, and it is the recommended model for use cases requiring high performance and accuracy. +Command R+ has been trained on a massive corpus of diverse texts in multiple languages, and can perform a wide array of text-generation tasks. You'll find it especially strong for complex RAG functionality, as well as workflows that lean on multi-step tool use to build agents.", + "indexSegmentId": "0", + "slug": "v2/changelog/command-r-is-a-scalable-llm-for-business", + "title": "Command R+ is a scalable LLM for business", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2023", + "January 1900", + ], + "content": "We're pleased to announce the release of Command R+, our newest and most performant large language model. Command R+ is optimized for conversational interaction and long-context tasks, and it is the recommended model for use cases requiring high performance and accuracy. +Command R+ has been trained on a massive corpus of diverse texts in multiple languages, and can perform a wide array of text-generation tasks. You'll find it especially strong for complex RAG functionality, as well as workflows that lean on multi-step tool use to build agents. +Multi-step Tool Use +Speaking of multi-step tool use, this functionality is now available for Command R+ models. +Multi-step tool use allows the model to call any number of tools in any sequence of steps, using the results from one tool call in a subsequent step until it has found a solution to a user's problem. This process allows the model to reason, perform dynamic actions, and quickly adapt on the basis of information coming from external sources.", + "indexSegmentId": "0", + "slug": "v2/changelog/command-r-is-a-scalable-llm-for-business", + "title": "Command R+ is a scalable LLM for business", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2023", + "title": "2023", + }, + { + "slug": "changelog/1", + "title": "January 1900", + }, + ], + "description": "Speaking of multi-step tool use, this functionality is now available for Command R+ models. +Multi-step tool use allows the model to call any number of tools in any sequence of steps, using the results from one tool call in a subsequent step until it has found a solution to a user's problem. This process allows the model to reason, perform dynamic actions, and quickly adapt on the basis of information coming from external sources.", + "indexSegmentId": "0", + "slug": "v2/changelog/command-r-is-a-scalable-llm-for-business#multi-step-tool-use", + "title": "Multi-step Tool Use", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2023", + "January 1900", + ], + "content": "Speaking of multi-step tool use, this functionality is now available for Command R+ models. +Multi-step tool use allows the model to call any number of tools in any sequence of steps, using the results from one tool call in a subsequent step until it has found a solution to a user's problem. This process allows the model to reason, perform dynamic actions, and quickly adapt on the basis of information coming from external sources.", + "indexSegmentId": "0", + "slug": "v2/changelog/command-r-is-a-scalable-llm-for-business#multi-step-tool-use", + "title": "Multi-step Tool Use", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, { "breadcrumbs": [ { @@ -65181,6 +65319,49 @@ The Co.classify endpoint now serves few-shot classification tasks using embeddin "slug": "", }, }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2022", + "title": "2022", + }, + { + "slug": "changelog/11", + "title": "November 1900", + }, + ], + "description": "Take control of your generative models with the new logit_bias parameter to guide token generation. + +Our Generative models have now the option to use the new logit_bias parameter to prevent the model from generating unwanted tokens or to incentivize it to include desired tokens. Logit bias is supported in all our default Generative models.", + "indexSegmentId": "0", + "slug": "v2/changelog/new-logit-bias-experimental-parameter", + "title": "New Logit Bias experimental parameter", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2022", + "November 1900", + ], + "content": "Our Generative models have now the option to use the new logit_bias parameter to prevent the model from generating unwanted tokens or to incentivize it to include desired tokens. Logit bias is supported in all our default Generative models.", + "indexSegmentId": "0", + "slug": "v2/changelog/new-logit-bias-experimental-parameter", + "title": "New Logit Bias experimental parameter", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, { "breadcrumbs": [ { @@ -65418,6 +65599,49 @@ Our new and improved xlarge has better generation quality and a 4x faster predic "slug": "", }, }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2022", + "title": "2022", + }, + { + "slug": "changelog/4", + "title": "April 1900", + }, + ], + "description": "The latest updates improve model stability and fix a bug for more effective generation presence and frequency penalties. + +Updated small, medium, and large models are more stable and resilient against abnormal inputs due to a FP16 quantization fix. We also fixed a bug in generation presence & frequency penalty, which will result in more effective penalties.", + "indexSegmentId": "0", + "slug": "v2/changelog/updated-small-medium-and-large-generation-models", + "title": "Updated Small, Medium, and Large Generation Models", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2022", + "April 1900", + ], + "content": "Updated small, medium, and large models are more stable and resilient against abnormal inputs due to a FP16 quantization fix. We also fixed a bug in generation presence & frequency penalty, which will result in more effective penalties.", + "indexSegmentId": "0", + "slug": "v2/changelog/updated-small-medium-and-large-generation-models", + "title": "Updated Small, Medium, and Large Generation Models", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, { "breadcrumbs": [ { @@ -65461,6 +65685,98 @@ Classification is now available via our classification endpoint. This endpoint i "slug": "", }, }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2022", + "title": "2022", + }, + { + "slug": "changelog/3", + "title": "March 1900", + }, + ], + "description": "Fine-tune models with your own data and leverage updated policies for powerful NLP solutions. + +Finetuning is Generally Available +You no longer need to wait for Full Access approval to build your own custom finetuned generation or representation model. Upload your dataset and start seeing even better performance for your specific task. +Policy Updates +The Cohere team continues to be focused on improving our products and features to enable our customers to build powerful NLP solutions. To help reflect some of the changes in our product development and research process, we have updated our Terms of Use, Privacy Policy, and click-through SaaS Agreement. Please carefully read and review these updates. By continuing to use Cohere’s services, you acknowledge that you have read, understood, and consent to all of the changes. If you have any questions or concerns about these updates, please contact us at support@cohere.ai.", + "indexSegmentId": "0", + "slug": "v2/changelog/finetuning-available-policy-updates", + "title": "Finetuning Available + Policy Updates", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2022", + "March 1900", + ], + "content": "Finetuning is Generally Available +You no longer need to wait for Full Access approval to build your own custom finetuned generation or representation model. Upload your dataset and start seeing even better performance for your specific task. +Policy Updates +The Cohere team continues to be focused on improving our products and features to enable our customers to build powerful NLP solutions. To help reflect some of the changes in our product development and research process, we have updated our Terms of Use, Privacy Policy, and click-through SaaS Agreement. Please carefully read and review these updates. By continuing to use Cohere’s services, you acknowledge that you have read, understood, and consent to all of the changes. If you have any questions or concerns about these updates, please contact us at support@cohere.ai.", + "indexSegmentId": "0", + "slug": "v2/changelog/finetuning-available-policy-updates", + "title": "Finetuning Available + Policy Updates", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + { + "slug": "changelog", + "title": "Release Notes", + }, + { + "slug": "changelog/2022", + "title": "2022", + }, + { + "slug": "changelog/3", + "title": "March 1900", + }, + ], + "description": "Try our new small, medium, and large generation models with improved performance from our high-quality dataset. + +We’ve shipped updated small, medium, and large generation models. You’ll find significant improvements in performance that come from our newly assembled high quality dataset.", + "indexSegmentId": "0", + "slug": "v2/changelog/new-improved-generation-models", + "title": "New & Improved Generation Models", + "type": "page-v4", + "version": { + "id": "v2 API", + "slug": "", + }, + }, + { + "breadcrumbs": [ + "Release Notes", + "2022", + "March 1900", + ], + "content": "We’ve shipped updated small, medium, and large generation models. You’ll find significant improvements in performance that come from our newly assembled high quality dataset.", + "indexSegmentId": "0", + "slug": "v2/changelog/new-improved-generation-models", + "title": "New & Improved Generation Models", + "type": "page-v3", + "version": { + "id": "v2 API", + "slug": "", + }, + }, { "breadcrumbs": [ { @@ -120464,322 +120780,6 @@ We can’t wait to see what you start building! Share your projects or find supp "slug": "v1", }, }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2024", - "title": "2024", - }, - { - "slug": "changelog/9", - "title": "September 1900", - }, - ], - "description": "Introducing our improved Command models are available on the Azure cloud computing platform. - -You'll recall that we released refreshed models of Command R and Command R+ in August. -Today, we're pleased to announce that these models are available on the Azure cloud computing platform! -You can find more information about using Cohere's Command models on Azure here.", - "indexSegmentId": "0", - "slug": "v2/changelog/v2-api-release", - "title": "The refreshed Command R and Command R+ models are now on Azure", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2024", - "September 1900", - ], - "content": "You'll recall that we released refreshed models of Command R and Command R+ in August. -Today, we're pleased to announce that these models are available on the Azure cloud computing platform! -You can find more information about using Cohere's Command models on Azure here.", - "indexSegmentId": "0", - "slug": "v2/changelog/v2-api-release", - "title": "The refreshed Command R and Command R+ models are now on Azure", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2023", - "title": "2023", - }, - { - "slug": "changelog/1", - "title": "January 1900", - }, - ], - "description": "Explore Command R+, Cohere's powerful language model, excelling in multi-step tool use and complex conversational AI tasks. - -We're pleased to announce the release of Command R+, our newest and most performant large language model. Command R+ is optimized for conversational interaction and long-context tasks, and it is the recommended model for use cases requiring high performance and accuracy. -Command R+ has been trained on a massive corpus of diverse texts in multiple languages, and can perform a wide array of text-generation tasks. You'll find it especially strong for complex RAG functionality, as well as workflows that lean on multi-step tool use to build agents.", - "indexSegmentId": "0", - "slug": "v2/changelog/command-model-nightly-available", - "title": "Command R+ is a scalable LLM for business", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2023", - "January 1900", - ], - "content": "We're pleased to announce the release of Command R+, our newest and most performant large language model. Command R+ is optimized for conversational interaction and long-context tasks, and it is the recommended model for use cases requiring high performance and accuracy. -Command R+ has been trained on a massive corpus of diverse texts in multiple languages, and can perform a wide array of text-generation tasks. You'll find it especially strong for complex RAG functionality, as well as workflows that lean on multi-step tool use to build agents. -Multi-step Tool Use -Speaking of multi-step tool use, this functionality is now available for Command R+ models. -Multi-step tool use allows the model to call any number of tools in any sequence of steps, using the results from one tool call in a subsequent step until it has found a solution to a user's problem. This process allows the model to reason, perform dynamic actions, and quickly adapt on the basis of information coming from external sources.", - "indexSegmentId": "0", - "slug": "v2/changelog/command-model-nightly-available", - "title": "Command R+ is a scalable LLM for business", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2023", - "title": "2023", - }, - { - "slug": "changelog/1", - "title": "January 1900", - }, - ], - "description": "Speaking of multi-step tool use, this functionality is now available for Command R+ models. -Multi-step tool use allows the model to call any number of tools in any sequence of steps, using the results from one tool call in a subsequent step until it has found a solution to a user's problem. This process allows the model to reason, perform dynamic actions, and quickly adapt on the basis of information coming from external sources.", - "indexSegmentId": "0", - "slug": "v2/changelog/command-model-nightly-available#multi-step-tool-use", - "title": "Multi-step Tool Use", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2023", - "January 1900", - ], - "content": "Speaking of multi-step tool use, this functionality is now available for Command R+ models. -Multi-step tool use allows the model to call any number of tools in any sequence of steps, using the results from one tool call in a subsequent step until it has found a solution to a user's problem. This process allows the model to reason, perform dynamic actions, and quickly adapt on the basis of information coming from external sources.", - "indexSegmentId": "0", - "slug": "v2/changelog/command-model-nightly-available#multi-step-tool-use", - "title": "Multi-step Tool Use", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2022", - "title": "2022", - }, - { - "slug": "changelog/11", - "title": "November 1900", - }, - ], - "description": "Take control of your generative models with the new logit_bias parameter to guide token generation. - -Our Generative models have now the option to use the new logit_bias parameter to prevent the model from generating unwanted tokens or to incentivize it to include desired tokens. Logit bias is supported in all our default Generative models.", - "indexSegmentId": "0", - "slug": "v2/changelog/coclassify-powered-by-our-representational-model-embeddings", - "title": "New Logit Bias experimental parameter", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2022", - "November 1900", - ], - "content": "Our Generative models have now the option to use the new logit_bias parameter to prevent the model from generating unwanted tokens or to incentivize it to include desired tokens. Logit bias is supported in all our default Generative models.", - "indexSegmentId": "0", - "slug": "v2/changelog/coclassify-powered-by-our-representational-model-embeddings", - "title": "New Logit Bias experimental parameter", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2022", - "title": "2022", - }, - { - "slug": "changelog/4", - "title": "April 1900", - }, - ], - "description": "The latest updates improve model stability and fix a bug for more effective generation presence and frequency penalties. - -Updated small, medium, and large models are more stable and resilient against abnormal inputs due to a FP16 quantization fix. We also fixed a bug in generation presence & frequency penalty, which will result in more effective penalties.", - "indexSegmentId": "0", - "slug": "v2/changelog/new-extremely-large-model", - "title": "Updated Small, Medium, and Large Generation Models", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2022", - "April 1900", - ], - "content": "Updated small, medium, and large models are more stable and resilient against abnormal inputs due to a FP16 quantization fix. We also fixed a bug in generation presence & frequency penalty, which will result in more effective penalties.", - "indexSegmentId": "0", - "slug": "v2/changelog/new-extremely-large-model", - "title": "Updated Small, Medium, and Large Generation Models", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2022", - "title": "2022", - }, - { - "slug": "changelog/3", - "title": "March 1900", - }, - ], - "description": "Fine-tune models with your own data and leverage updated policies for powerful NLP solutions. - -Finetuning is Generally Available -You no longer need to wait for Full Access approval to build your own custom finetuned generation or representation model. Upload your dataset and start seeing even better performance for your specific task. -Policy Updates -The Cohere team continues to be focused on improving our products and features to enable our customers to build powerful NLP solutions. To help reflect some of the changes in our product development and research process, we have updated our Terms of Use, Privacy Policy, and click-through SaaS Agreement. Please carefully read and review these updates. By continuing to use Cohere’s services, you acknowledge that you have read, understood, and consent to all of the changes. If you have any questions or concerns about these updates, please contact us at support@cohere.ai.", - "indexSegmentId": "0", - "slug": "v2/changelog/classification-endpoint", - "title": "Finetuning Available + Policy Updates", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2022", - "March 1900", - ], - "content": "Finetuning is Generally Available -You no longer need to wait for Full Access approval to build your own custom finetuned generation or representation model. Upload your dataset and start seeing even better performance for your specific task. -Policy Updates -The Cohere team continues to be focused on improving our products and features to enable our customers to build powerful NLP solutions. To help reflect some of the changes in our product development and research process, we have updated our Terms of Use, Privacy Policy, and click-through SaaS Agreement. Please carefully read and review these updates. By continuing to use Cohere’s services, you acknowledge that you have read, understood, and consent to all of the changes. If you have any questions or concerns about these updates, please contact us at support@cohere.ai.", - "indexSegmentId": "0", - "slug": "v2/changelog/classification-endpoint", - "title": "Finetuning Available + Policy Updates", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - { - "slug": "changelog", - "title": "Release Notes", - }, - { - "slug": "changelog/2022", - "title": "2022", - }, - { - "slug": "changelog/3", - "title": "March 1900", - }, - ], - "description": "Try our new small, medium, and large generation models with improved performance from our high-quality dataset. - -We’ve shipped updated small, medium, and large generation models. You’ll find significant improvements in performance that come from our newly assembled high quality dataset.", - "indexSegmentId": "0", - "slug": "v2/changelog/classification-endpoint", - "title": "New & Improved Generation Models", - "type": "page-v4", - "version": { - "id": "v2 API", - "slug": "", - }, - }, - { - "breadcrumbs": [ - "Release Notes", - "2022", - "March 1900", - ], - "content": "We’ve shipped updated small, medium, and large generation models. You’ll find significant improvements in performance that come from our newly assembled high quality dataset.", - "indexSegmentId": "0", - "slug": "v2/changelog/classification-endpoint", - "title": "New & Improved Generation Models", - "type": "page-v3", - "version": { - "id": "v2 API", - "slug": "", - }, - }, { "breadcrumbs": [ "Endpoints",