feat: enable Nemotron 3.5 Lightning defaults

.agents/skills/aiq-configure-workflow/assets/config-scaffold.yml

@@ -17,25 +17,29 @@ general:
   # Web/API: copy general.front_end from config_web_default_llamaindex.yml
 
 llms:
-  nemotron_llm_intent:
+  nemotron_lightning_intent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.5
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.1
     top_p: 0.9
-    max_tokens: 4096
+    max_tokens: 1024
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
-      enable_thinking: true
+      enable_thinking: false
 
-  nemotron_super_llm:
+  nemotron_lightning_agent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.7
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.2
     top_p: 0.7
-    max_tokens: 65536
+    max_tokens: 8192
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
       enable_thinking: true
 
@@ -68,11 +72,11 @@ functions:
 
   intent_classifier:
     _type: intent_classifier
-    llm: nemotron_llm_intent
+    llm: nemotron_lightning_intent_llm
 
   shallow_research_agent:
     _type: shallow_research_agent
-    llm: nemotron_super_llm
+    llm: nemotron_lightning_agent_llm
 
   deep_research_agent:
     _type: deep_research_agent

.agents/skills/aiq-configure-workflow/references/config-profiles.md

@@ -17,7 +17,7 @@ the file from `configs/`. For the full nine-profile table, see
 3. **Model family**
    - Nemotron → most profiles
    - GPT-5.2 orchestration/planning/writing → `config_frontier_models.yml` (`OPENAI_API_KEY`)
-   - Default split → existing Super profiles for intent/shallow, with Ultra for deep research and a larger writer budget
+   - Default split → Nemotron 3.5 Lightning for intent/shallow; Ultra for clarification/deep research, with a larger writer budget
 
 4. **Optional features** — copy blocks from:
    - Guardrails → `config_web_default_guardrails.yml`

README.md

@@ -108,8 +108,8 @@ The checked-in default CLI and web profiles use these core components:
 
 - [NVIDIA NeMo Agent Toolkit 1.8.0](https://docs.nvidia.com/nemo/agent-toolkit/latest/)
 - [LangChain Deep Agents](https://docs.langchain.com/oss/python/deepagents/overview) 0.6.5 or newer
-- [NVIDIA Nemotron 3 Super](https://build.nvidia.com/nvidia/nemotron-3-super-120b-a12b) for intent classification and shallow research in the default profiles
-- [NVIDIA Nemotron 3 Ultra](https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b) for clarification and every deep-research role in the default profiles
+- NVIDIA Nemotron 3.5 Lightning for intent classification and shallow research in the default profiles
+- [NVIDIA Nemotron 3 Ultra](https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b) for clarification and every deep-research role
 - [Google Gemma 4 31B IT](https://build.nvidia.com/google/gemma-4-31b-it) (document summary, if used)
 - [NVIDIA Nemotron 3 Embed 1B](https://build.nvidia.com/nvidia/nemotron-3-embed-1b) (embedding model for the knowledge layer, if used)
 - [NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning](https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning) (vision-language model for the LlamaIndex knowledge layer, if used)
@@ -152,7 +152,7 @@ When using [NVIDIA API Catalog](https://build.nvidia.com/) (the default), infere
 
 | Component | Default Model | Self-Hosted Hardware Reference |
 |-----------|---------------|-------------------------------|
-| LLM (intent classifier, shallow researcher) | `nvidia/nemotron-3-super-120b-a12b` | [Nemotron 3 Super](https://build.nvidia.com/nvidia/nemotron-3-super-120b-a12b) |
+| LLM (intent classifier, shallow researcher) | `nvidia/nemotron-3.5-lightning-30b-a3b` | [Nemotron 3.5 Lightning](https://build.nvidia.com/nvidia/nemotron-3.5-lightning-30b-a3b/modelcard) |
 | LLM (clarifier and all deep-research roles) | `nvidia/nemotron-3-ultra-550b-a55b` | [Nemotron 3 Ultra](https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b) |
 | Document summary (optional) | `google/gemma-4-31b-it` | [Gemma 4 31B IT](https://build.nvidia.com/google/gemma-4-31b-it) |
 | Text embedding | `nvidia/nemotron-3-embed-1b` | [NeMo Retriever embedding support matrix](https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html) |
@@ -289,17 +289,17 @@ The `configs/` directory holds YAML workflow configs that define agents, tools,
 
 | Config | Models | Description |
 |--------|--------|-------------|
-| `config_cli_default.yml` | Nemotron 3 Super/Ultra | CLI chat pipeline with Tavily and clarification; no knowledge backend. Paper search is a commented opt-in. |
-| `config_web_default_llamaindex.yml` | Nemotron 3 Super/Ultra; Gemma 4 summary | Default web/API chat pipeline with LlamaIndex/ChromaDB and Tavily. Paper search is commented out. |
-| `config_web_frag.yml` | Nemotron 3 Super/Ultra | Web/API and Helm base with Foundational RAG plus Tavily. Requires separately deployed RAG query and ingestion services. |
-| `config_web_opensearch.yml` | Nemotron 3 Super/Ultra; Nemotron 3 Embed | Web/API with built-in OpenSearch knowledge retrieval plus Tavily; supports self-hosted, `es`, and `aoss` authentication modes. |
-| `config_web_azure_ai_search.yml` | Nemotron 3 Super/Ultra; Nemotron 3 Embed | Web/API with Azure AI Search knowledge retrieval plus Tavily; supports API-key and Azure identity authentication. |
-| `config_frontier_models.yml` | GPT Sol/Luna; Nemotron 3 Super; Gemma 4 summary | LlamaIndex profile using GPT Sol/Luna for deep research and Super for intent/shallow. Requires `OPENAI_API_KEY` and `NVIDIA_API_KEY`. |
-| `config_web_default_guardrails.yml` | Nemotron 3 Super/Ultra; Gemma 4 summary | LlamaIndex profile with workflow Guardrails explicitly attached, shallow-agent Guardrails dynamically attached through `workflow_functions`, and async deep-agent Guardrails applied by the AI-Q runner from the same target configuration. |
-| `config_web_frag_mcp_auth.yml` | Nemotron 3 Super/Ultra | Foundational RAG plus an opt-in protected per-user OAuth MCP source example. Requires a real MCP endpoint and shared token store. |
+| `config_cli_default.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra | CLI chat pipeline with Tavily and clarification; no knowledge backend. Paper search is a commented opt-in. |
+| `config_web_default_llamaindex.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra; Gemma 4 summary | Default web/API chat pipeline with LlamaIndex/ChromaDB and Tavily. Paper search is commented out. |
+| `config_web_frag.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra | Web/API and Helm base with Foundational RAG plus Tavily. Requires separately deployed RAG query and ingestion services. |
+| `config_web_opensearch.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra; Nemotron 3 Embed | Web/API with built-in OpenSearch knowledge retrieval plus Tavily; supports self-hosted, `es`, and `aoss` authentication modes. |
+| `config_web_azure_ai_search.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra; Nemotron 3 Embed | Web/API with Azure AI Search knowledge retrieval plus Tavily; supports API-key and Azure identity authentication. |
+| `config_frontier_models.yml` | GPT Sol/Luna; Nemotron 3.5 Lightning; Gemma 4 summary | LlamaIndex profile using GPT Sol/Luna for deep research and Nemotron 3.5 Lightning for intent/shallow. Requires `OPENAI_API_KEY` and `NVIDIA_API_KEY`. |
+| `config_web_default_guardrails.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra; Gemma 4 summary | LlamaIndex profile with workflow Guardrails explicitly attached, shallow-agent Guardrails dynamically attached through `workflow_functions`, and async deep-agent Guardrails applied by the AI-Q runner from the same target configuration. |
+| `config_web_frag_mcp_auth.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra | Foundational RAG plus an opt-in protected per-user OAuth MCP source example. Requires a real MCP endpoint and shared token store. |
 | `config_domain_routing_and_skills.yml` | Nemotron 3 Ultra; Gemma 4 summary | Direct deep-research profile with domain routing, DuckDuckGo news, Polymarket, enabled Serper paper search, LlamaIndex, built-in skills, and a fresh per-job Modal sandbox. |
-| `config_openshell.yml` | Nemotron 3 Super/Ultra; Gemma 4 summary | Experimental web/API skills profile with artifact capture, fail-closed policy attestation, and one OpenShell sandbox per deep-research job. |
-| `config_mcp.yml` | Nemotron 3 Super/Ultra | Standalone MCP server. Public NIM + Tavily research with PostgreSQL-backed stateless submit/poll/report. Requires `NVIDIA_API_KEY`, `TAVILY_API_KEY`, and `AIQ_CHECKPOINT_DB`. |
+| `config_openshell.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra; Gemma 4 summary | Experimental web/API skills profile with artifact capture, fail-closed policy attestation, and one OpenShell sandbox per deep-research job. |
+| `config_mcp.yml` | Nemotron 3.5 Lightning; Nemotron 3 Ultra | Standalone MCP server. Public NIM + Tavily research with PostgreSQL-backed stateless submit/poll/report. Requires `NVIDIA_API_KEY`, `TAVILY_API_KEY`, and `AIQ_CHECKPOINT_DB`. |
 
 ## Ways to Run the Agents
 

configs/config_cli_default.yml

@@ -16,25 +16,29 @@ general:
     #     project: nvidia-aiq
 
 llms:
-  nemotron_llm_intent:
+  nemotron_lightning_intent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.5
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.1
     top_p: 0.9
-    max_tokens: 4096
+    max_tokens: 1024
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
-      enable_thinking: true
+      enable_thinking: false
 
-  nemotron_super_llm:
+  nemotron_lightning_agent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.7
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.2
     top_p: 0.7
-    max_tokens: 65536
+    max_tokens: 8192
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
       enable_thinking: true
 
@@ -115,7 +119,7 @@ functions:
   # =========================================================================
   intent_classifier:
     _type: intent_classifier
-    llm: nemotron_llm_intent
+    llm: nemotron_lightning_intent_llm
     # tools: omitted -> inherits all from data_source_registry
     # exclude_tools: []
     # llm_timeout: 90  # optional; seconds for intent LLM call (default 90)
@@ -131,7 +135,7 @@ functions:
 
   shallow_research_agent:
     _type: shallow_research_agent
-    llm: nemotron_super_llm
+    llm: nemotron_lightning_agent_llm
     # tools: omitted -> inherits all from data_source_registry
     exclude_tools:
       - advanced_web_search_tool

configs/config_frontier_models.yml

@@ -51,25 +51,29 @@ general:
         - "*"
 
 llms:
-  nemotron_llm_intent:
+  nemotron_lightning_intent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.5
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.1
     top_p: 0.9
-    max_tokens: 4096
+    max_tokens: 1024
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
-      enable_thinking: true
+      enable_thinking: false
 
-  nemotron_super_llm:
+  nemotron_lightning_agent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.1
-    top_p: 0.3
-    max_tokens: 16384
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.2
+    top_p: 0.7
+    max_tokens: 8192
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
       enable_thinking: true
 
@@ -137,7 +141,7 @@ functions:
 
   intent_classifier:
     _type: intent_classifier
-    llm: nemotron_llm_intent
+    llm: nemotron_lightning_intent_llm
     verbose: true
     tools:
       - web_search_tool
@@ -156,7 +160,7 @@ functions:
 
   shallow_research_agent:
     _type: shallow_research_agent
-    llm: nemotron_super_llm
+    llm: nemotron_lightning_agent_llm
     verbose: true
     tools:
       - web_search_tool

configs/config_mcp.yml

@@ -26,25 +26,29 @@ general:
         level: INFO
 
 llms:
-  nemotron_llm_intent:
+  nemotron_lightning_intent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.5
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.1
     top_p: 0.9
-    max_tokens: 4096
+    max_tokens: 1024
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
-      enable_thinking: true
+      enable_thinking: false
 
-  nemotron_super_llm:
+  nemotron_lightning_agent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.7
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.2
     top_p: 0.7
-    max_tokens: 65536
+    max_tokens: 8192
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
       enable_thinking: true
 
@@ -95,7 +99,7 @@ functions:
 
   intent_classifier:
     _type: intent_classifier
-    llm: nemotron_llm_intent
+    llm: nemotron_lightning_intent_llm
 
   # Defined for workflow compatibility but disabled at the workflow boundary.
   clarifier_agent:
@@ -107,7 +111,7 @@ functions:
 
   shallow_research_agent:
     _type: shallow_research_agent
-    llm: nemotron_super_llm
+    llm: nemotron_lightning_agent_llm
     exclude_tools:
       - advanced_web_search_tool
     max_llm_turns: 10

configs/config_openshell.yml

@@ -28,25 +28,29 @@ general:
         - "*"
 
 llms:
-  nemotron_llm_intent:
+  nemotron_lightning_intent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.5
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.1
     top_p: 0.9
-    max_tokens: 4096
+    max_tokens: 1024
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
-      enable_thinking: true
+      enable_thinking: false
 
-  nemotron_super_llm:
+  nemotron_lightning_agent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.7
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.2
     top_p: 0.7
-    max_tokens: 65536
+    max_tokens: 8192
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
       enable_thinking: true
 
@@ -120,7 +124,7 @@ functions:
 
   intent_classifier:
     _type: intent_classifier
-    llm: nemotron_llm_intent
+    llm: nemotron_lightning_intent_llm
     verbose: true
 
   clarifier_agent:
@@ -132,7 +136,7 @@ functions:
 
   shallow_research_agent:
     _type: shallow_research_agent
-    llm: nemotron_super_llm
+    llm: nemotron_lightning_agent_llm
     exclude_tools:
       - advanced_web_search_tool
     verbose: true

configs/config_web_azure_ai_search.yml

@@ -47,25 +47,29 @@ general:
         - "*"
 
 llms:
-  nemotron_llm_intent:
+  nemotron_lightning_intent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.5
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.1
     top_p: 0.9
-    max_tokens: 4096
+    max_tokens: 1024
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
-      enable_thinking: true
+      enable_thinking: false
 
-  nemotron_super_llm:
+  nemotron_lightning_agent_llm:
     _type: nim
-    model_name: nvidia/nemotron-3-super-120b-a12b
+    model_name: nvidia/nemotron-3.5-lightning-30b-a3b
     base_url: "https://integrate.api.nvidia.com/v1"
-    temperature: 0.7
+    api_key: ${NVIDIA_API_KEY}
+    temperature: 0.2
     top_p: 0.7
-    max_tokens: 65536
+    max_tokens: 8192
     num_retries: 5
+    parallel_tool_calls: false
     chat_template_kwargs:
       enable_thinking: true
 
@@ -187,7 +191,7 @@ functions:
   # =========================================================================
   intent_classifier:
     _type: intent_classifier
-    llm: nemotron_llm_intent
+    llm: nemotron_lightning_intent_llm
     # tools: omitted -> inherits all from data_source_registry
     # exclude_tools: []
     verbose: true
@@ -205,7 +209,7 @@ functions:
 
   shallow_research_agent:
     _type: shallow_research_agent
-    llm: nemotron_super_llm
+    llm: nemotron_lightning_agent_llm
     # tools: omitted -> inherits all from data_source_registry
     exclude_tools:                     # Remove advanced variant; shallow uses web_search_tool
       - advanced_web_search_tool