Merge pull request #23 from Not-Diamond/release-please--branches--main--changes--next

release: 1.0.0-rc15

Author: dragos199993

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "1.0.0-rc14"
+  ".": "1.0.0-rc15"
 }

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 11
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/not-diamond-dragos199993%2Fnotdiamond-ce07c2efae541713a6bc9c540111adb2b9ff9fd472f77338d851cfb4b92239f4.yml
-openapi_spec_hash: 9536c08e10a84315ce7555faff6c48dd
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/not-diamond-dragos199993%2Fnotdiamond-726179b7b0c90f2715aa3173af7dc889f80d79b376ba6bb58cb9481965f42d5a.yml
+openapi_spec_hash: cd57c158ad6a480fef03dd3e2d4e2dfa
 config_hash: 11ee9bddc4b7df5f656add53f8f1ed67

CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 1.0.0-rc15 (2025-11-14)
+
+Full Changelog: [v1.0.0-rc14...v1.0.0-rc15](https://github.com/Not-Diamond/not-diamond-python/compare/v1.0.0-rc14...v1.0.0-rc15)
+
+### Features
+
+* **api:** api update ([12c16a2](https://github.com/Not-Diamond/not-diamond-python/commit/12c16a218d211c47aaad82ad672f668a27715edf))
+
 ## 1.0.0-rc14 (2025-11-14)
 
 Full Changelog: [v1.0.0-rc13...v1.0.0-rc14](https://github.com/Not-Diamond/not-diamond-python/compare/v1.0.0-rc13...v1.0.0-rc14)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "notdiamond"
-version = "1.0.0-rc14"
+version = "1.0.0-rc15"
 description = "The official Python library for the notdiamond API"
 dynamic = ["readme"]
 license = "Apache-2.0"

src/notdiamond/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "notdiamond"
-__version__ = "1.0.0-rc14"  # x-release-please-version
+__version__ = "1.0.0-rc15"  # x-release-please-version

src/notdiamond/resources/model_router.py

@@ -68,7 +68,7 @@ def select_model(
         algorithm.
 
         This endpoint analyzes your messages and returns the best-suited model from your
-        specified providers. The router considers factors like query complexity, model
+        specified models. The router considers factors like query complexity, model
         capabilities, cost, and latency based on your preferences.
 
         **Key Features:**
@@ -77,14 +77,13 @@ def select_model(
         - Support for custom routers trained on your evaluation data
         - Optional cost/latency optimization
         - Function calling support for compatible models
-        - Privacy-preserving content hashing
 
         **Usage:**
 
         1. Pass your messages in OpenAI format (array of objects with 'role' and
            'content')
         2. Specify which LLM providers you want to route between
-        3. Optionally provide a preference_id for personalized routing
+        3. Optionally provide a preference_id to use a custom router that you've trained
         4. Receive a recommended model and session_id
         5. Use the session_id to submit feedback and improve routing
 
@@ -202,7 +201,7 @@ async def select_model(
         algorithm.
 
         This endpoint analyzes your messages and returns the best-suited model from your
-        specified providers. The router considers factors like query complexity, model
+        specified models. The router considers factors like query complexity, model
         capabilities, cost, and latency based on your preferences.
 
         **Key Features:**
@@ -211,14 +210,13 @@ async def select_model(
         - Support for custom routers trained on your evaluation data
         - Optional cost/latency optimization
         - Function calling support for compatible models
-        - Privacy-preserving content hashing
 
         **Usage:**
 
         1. Pass your messages in OpenAI format (array of objects with 'role' and
            'content')
         2. Specify which LLM providers you want to route between
-        3. Optionally provide a preference_id for personalized routing
+        3. Optionally provide a preference_id to use a custom router that you've trained
         4. Receive a recommended model and session_id
         5. Use the session_id to submit feedback and improve routing
 

src/notdiamond/resources/prompt_adaptation.py

@@ -73,23 +73,22 @@ def adapt(
         Adapt your prompt from one LLM to work optimally across different target LLMs.
 
         This endpoint automatically optimizes your prompt (system prompt + user message
-        template) to achieve better performance when switching between different
-        language models. Each model has unique characteristics, and what works well for
-        GPT-4 might not work as well for Claude or Gemini.
+        template) to improve accuracy on your use case across various models. Each model
+        has unique characteristics, and what works well for GPT-5 might not work as well
+        for Claude or Gemini.
 
         **How Prompt Adaptation Works:**
 
-        1. You provide your current prompt optimized for an origin model
-        2. You specify target models you want to adapt to
+        1. You provide your current prompt and optionally your current origin model
+        2. You specify the target models you want to adapt your prompt to
         3. You provide evaluation examples (golden records) with expected answers
         4. The system runs optimization to find the best prompt for each target model
         5. You receive adapted prompts that perform well on your target models
 
         **Evaluation Metrics:** Choose either a standard metric or provide custom
         evaluation:
 
-        - **Standard metrics**: LLMaaJ:SQL, LLMaaJ:Sem_Sim_1/3/10 (semantic similarity),
-          JSON_Match
+        - **Standard metrics**: LLMaaJ:Sem_Sim_1 (semantic similarity), JSON_Match
         - **Custom evaluation**: Provide evaluation_config with your own LLM judge,
           prompt, and cutoff
 
@@ -106,21 +105,6 @@ def adapt(
         - Time depends on: number of target models, dataset size, model availability
         - Use the returned adaptation_run_id to check status and retrieve results
 
-        **Subscription Tiers:**
-
-        - Free: 1 target model
-        - Starter: 3 target models
-        - Startup: 5 target models
-        - Enterprise: 10 target models
-
-        **Best Practices:**
-
-        1. Use diverse, representative examples from your production workload
-        2. Include examples for best results (25 minimum)
-        3. Ensure consistent evaluation across all examples
-        4. Test both train_goldens and test_goldens split for validation
-        5. Use the same model versions you'll use in production
-
         **Example Workflow:**
 
         ```
@@ -232,15 +216,6 @@ def get_adapt_results(
         3. Apply the optimized prompts when calling the respective target models
         4. Compare pre/post optimization scores to see improvement
 
-        **Evaluation Scores:**
-
-        - Scores range from 0-10 (higher is better)
-        - Compare origin_model score with target_models pre_optimization_score for
-          baseline
-        - Compare pre_optimization_score with post_optimization_score to see improvement
-          from adaptation
-        - Typical improvements range from 5-30% on evaluation metrics
-
         **Status Handling:**
 
         - If adaptation is still processing, target model results will have
@@ -366,7 +341,6 @@ def get_costs(
         - Total cost across all models used in the adaptation
         - Individual usage records with provider, model, tokens, and costs
         - Timestamps for each LLM request
-        - Task type (e.g., optimization, evaluation)
 
         **Access Control:**
 
@@ -438,23 +412,22 @@ async def adapt(
         Adapt your prompt from one LLM to work optimally across different target LLMs.
 
         This endpoint automatically optimizes your prompt (system prompt + user message
-        template) to achieve better performance when switching between different
-        language models. Each model has unique characteristics, and what works well for
-        GPT-4 might not work as well for Claude or Gemini.
+        template) to improve accuracy on your use case across various models. Each model
+        has unique characteristics, and what works well for GPT-5 might not work as well
+        for Claude or Gemini.
 
         **How Prompt Adaptation Works:**
 
-        1. You provide your current prompt optimized for an origin model
-        2. You specify target models you want to adapt to
+        1. You provide your current prompt and optionally your current origin model
+        2. You specify the target models you want to adapt your prompt to
         3. You provide evaluation examples (golden records) with expected answers
         4. The system runs optimization to find the best prompt for each target model
         5. You receive adapted prompts that perform well on your target models
 
         **Evaluation Metrics:** Choose either a standard metric or provide custom
         evaluation:
 
-        - **Standard metrics**: LLMaaJ:SQL, LLMaaJ:Sem_Sim_1/3/10 (semantic similarity),
-          JSON_Match
+        - **Standard metrics**: LLMaaJ:Sem_Sim_1 (semantic similarity), JSON_Match
         - **Custom evaluation**: Provide evaluation_config with your own LLM judge,
           prompt, and cutoff
 
@@ -471,21 +444,6 @@ async def adapt(
         - Time depends on: number of target models, dataset size, model availability
         - Use the returned adaptation_run_id to check status and retrieve results
 
-        **Subscription Tiers:**
-
-        - Free: 1 target model
-        - Starter: 3 target models
-        - Startup: 5 target models
-        - Enterprise: 10 target models
-
-        **Best Practices:**
-
-        1. Use diverse, representative examples from your production workload
-        2. Include examples for best results (25 minimum)
-        3. Ensure consistent evaluation across all examples
-        4. Test both train_goldens and test_goldens split for validation
-        5. Use the same model versions you'll use in production
-
         **Example Workflow:**
 
         ```
@@ -597,15 +555,6 @@ async def get_adapt_results(
         3. Apply the optimized prompts when calling the respective target models
         4. Compare pre/post optimization scores to see improvement
 
-        **Evaluation Scores:**
-
-        - Scores range from 0-10 (higher is better)
-        - Compare origin_model score with target_models pre_optimization_score for
-          baseline
-        - Compare pre_optimization_score with post_optimization_score to see improvement
-          from adaptation
-        - Typical improvements range from 5-30% on evaluation metrics
-
         **Status Handling:**
 
         - If adaptation is still processing, target model results will have
@@ -731,7 +680,6 @@ async def get_costs(
         - Total cost across all models used in the adaptation
         - Individual usage records with provider, model, tokens, and costs
         - Timestamps for each LLM request
-        - Task type (e.g., optimization, evaluation)
 
         **Access Control:**