Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 19 additions & 19 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -82,6 +82,7 @@ Purpose: Demonstrate NeMo Gym patterns and concepts.
<!-- START_EXAMPLE_ONLY_SERVERS_TABLE -->
| Name | Demonstrates | Config | README |
| ---------------- | ------------------------------------ | ------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------- |
| Multi Step | Instruction_Following example | <a href='resources_servers/example_multi_step/configs/example_multi_step.yaml'>example_multi_step.yaml</a> | <a href='resources_servers/example_multi_step/README.md'>README</a> |
| Simple Weather | Basic single-step tool calling | <a href='resources_servers/example_simple_weather/configs/simple_weather.yaml'>simple_weather.yaml</a> | <a href='resources_servers/example_simple_weather/README.md'>README</a> |
| Stateful Counter | Session state management (in-memory) | <a href='resources_servers/example_stateful_counter/configs/stateful_counter.yaml'>stateful_counter.yaml</a> | <a href='resources_servers/example_stateful_counter/README.md'>README</a> |
<!-- END_EXAMPLE_ONLY_SERVERS_TABLE -->
Expand All @@ -94,25 +95,24 @@ Purpose: Training-ready environments with curated datasets.
> Each resource server includes example data, configuration files, and tests. See each server's README for details.

<!-- START_TRAINING_SERVERS_TABLE -->
| Domain | Resource Server | Train | Validation | Verified | Config | License |
| --------------------- | --------------------- | ----- | ---------- | ------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------- |
| coding | Comp Coding | ✓ | ✓ | <a href='https://wandb.ai/nvidia/bxyu-nemo-gym-rl-integration-20250926/runs/54uzarwq'>✓</a> | <a href='resources_servers/comp_coding/configs/comp_coding.yaml'>comp_coding.yaml</a> | Apache 2.0 |
| math | Library Judge Math | ✓ | ✓ | <a href='https://wandb.ai/nvidia/bxyu-nemo-gym-rl-integration/runs/5d2a98ix'>✓</a> | <a href='resources_servers/library_judge_math/configs/bytedtsinghua_dapo17k.yaml'>bytedtsinghua_dapo17k.yaml</a> | Apache 2.0 |
| agent | Google Search | ✓ | - | - | <a href='resources_servers/google_search/configs/google_search.yaml'>google_search.yaml</a> | Apache 2.0 |
| agent | Multiverse Math Hard | ✓ | - | - | <a href='resources_servers/multiverse_math_hard/configs/multiverse_math_hard.yaml'>multiverse_math_hard.yaml</a> | Apache 2.0 |
| agent | Workbench | ✓ | ✓ | - | <a href='resources_servers/workbench/configs/workbench.yaml'>workbench.yaml</a> | Apache 2.0 |
| coding | Mini Swe Resource | ✓ | ✓ | - | <a href='resources_servers/mini_swe_resource/configs/mini_swe_resource.yaml'>mini_swe_resource.yaml</a> | MIT |
| instruction_following | Instruction Following | ✓ | - | - | <a href='resources_servers/instruction_following/configs/instruction_following.yaml'>instruction_following.yaml</a> | Apache 2.0 |
| instruction_following | Multineedle | ✓ | ✓ | - | <a href='resources_servers/multineedle/configs/multineedle.yaml'>multineedle.yaml</a> | Apache 2.0 |
| instruction_following | Structured Outputs | ✓ | ✓ | - | <a href='resources_servers/structured_outputs/configs/structured_outputs_json.yaml'>structured_outputs_json.yaml</a> | Apache 2.0 |
| knowledge | Equivalence Llm Judge | - | - | - | <a href='resources_servers/equivalence_llm_judge/configs/equivalence_llm_judge.yaml'>equivalence_llm_judge.yaml</a> | - |
| knowledge | Mcqa | ✓ | - | - | <a href='resources_servers/mcqa/configs/mcqa.yaml'>mcqa.yaml</a> | Apache 2.0 |
| math | Library Judge Math | - | ✓ | - | <a href='resources_servers/library_judge_math/configs/dapo17k_trajectory_collection.yaml'>dapo17k_trajectory_collection.yaml</a> | - |
| math | Library Judge Math | ✓ | ✓ | - | <a href='resources_servers/library_judge_math/configs/dapo17k.yaml'>dapo17k.yaml</a> | Apache 2.0 |
| math | Library Judge Math | ✓ | ✓ | - | <a href='resources_servers/library_judge_math/configs/dapo17k_filtered_qwen330ba3binstruct.yaml'>dapo17k_filtered_qwen330ba3binstruct.yaml</a> | Apache 2.0 |
| math | Library Judge Math | ✓ | ✓ | - | <a href='resources_servers/library_judge_math/configs/library_judge_math.yaml'>library_judge_math.yaml</a> | Creative Commons Attribution 4.0 International |
| math | Library Judge Math | ✓ | ✓ | - | <a href='resources_servers/library_judge_math/configs/math_stack_overflow.yaml'>math_stack_overflow.yaml</a> | Creative Commons Attribution-ShareAlike 4.0 International |
| math | Python Math Exec | ✓ | - | - | <a href='resources_servers/python_math_exec/configs/python_math_exec.yaml'>python_math_exec.yaml</a> | Apache 2.0 |
| Domain | Resource Server | Train | Validation | Verified | Config | License |
| --------------------- | -------------------------- | ----- | ---------- | ------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------- |
| coding | Code Gen | ✓ | ✓ | <a href='https://wandb.ai/nvidia/bxyu-nemo-gym-rl-integration-20250926/runs/54uzarwq'>✓</a> | <a href='resources_servers/code_gen/configs/code_gen.yaml'>code_gen.yaml</a> | Apache 2.0 |
| math | Math With Judge | ✓ | ✓ | <a href='https://wandb.ai/nvidia/bxyu-nemo-gym-rl-integration/runs/5d2a98ix'>✓</a> | <a href='resources_servers/math_with_judge/configs/bytedtsinghua_dapo17k.yaml'>bytedtsinghua_dapo17k.yaml</a> | Apache 2.0 |
| agent | Google Search | ✓ | - | - | <a href='resources_servers/google_search/configs/google_search.yaml'>google_search.yaml</a> | Apache 2.0 |
| agent | Math Advanced Calculations | ✓ | - | - | <a href='resources_servers/math_advanced_calculations/configs/math_advanced_calculations.yaml'>math_advanced_calculations.yaml</a> | Apache 2.0 |
| agent | Workplace Assistant | ✓ | ✓ | - | <a href='resources_servers/workplace_assistant/configs/workplace_assistant.yaml'>workplace_assistant.yaml</a> | Apache 2.0 |
| coding | Mini Swe Agent | ✓ | ✓ | - | <a href='resources_servers/mini_swe_agent/configs/mini_swe_agent.yaml'>mini_swe_agent.yaml</a> | MIT |
| instruction_following | Instruction Following | ✓ | - | - | <a href='resources_servers/instruction_following/configs/instruction_following.yaml'>instruction_following.yaml</a> | Apache 2.0 |
| instruction_following | Structured Outputs | ✓ | ✓ | - | <a href='resources_servers/structured_outputs/configs/structured_outputs_json.yaml'>structured_outputs_json.yaml</a> | Apache 2.0 |
| knowledge | Equivalence Llm Judge | - | - | - | <a href='resources_servers/equivalence_llm_judge/configs/equivalence_llm_judge.yaml'>equivalence_llm_judge.yaml</a> | - |
| knowledge | Mcqa | ✓ | - | - | <a href='resources_servers/mcqa/configs/mcqa.yaml'>mcqa.yaml</a> | Apache 2.0 |
| math | Math With Code | ✓ | - | - | <a href='resources_servers/math_with_code/configs/math_with_code.yaml'>math_with_code.yaml</a> | Apache 2.0 |
| math | Math With Judge | - | ✓ | - | <a href='resources_servers/math_with_judge/configs/dapo17k_trajectory_collection.yaml'>dapo17k_trajectory_collection.yaml</a> | - |
| math | Math With Judge | ✓ | ✓ | - | <a href='resources_servers/math_with_judge/configs/dapo17k.yaml'>dapo17k.yaml</a> | Apache 2.0 |
| math | Math With Judge | ✓ | ✓ | - | <a href='resources_servers/math_with_judge/configs/dapo17k_filtered_qwen330ba3binstruct.yaml'>dapo17k_filtered_qwen330ba3binstruct.yaml</a> | Apache 2.0 |
| math | Math With Judge | ✓ | ✓ | - | <a href='resources_servers/math_with_judge/configs/math_stack_overflow.yaml'>math_stack_overflow.yaml</a> | Creative Commons Attribution-ShareAlike 4.0 International |
| math | Math With Judge | ✓ | ✓ | - | <a href='resources_servers/math_with_judge/configs/math_with_judge.yaml'>math_with_judge.yaml</a> | Creative Commons Attribution 4.0 International |
<!-- END_TRAINING_SERVERS_TABLE -->


6 changes: 3 additions & 3 deletions docs/about/concepts/configuration-system.md
Original file line number Diff line number Diff line change
Expand Up @@ -189,7 +189,7 @@ ng_run "+config_paths=[prod-config.yaml]" +policy_model_name=gpt-4o-2024-11-20
```bash
# Test with math resources
ng_run "+config_paths=[base.yaml]" \
+simple_agent.responses_api_agents.simple_agent.resources_server.name=library_judge_math
+simple_agent.responses_api_agents.simple_agent.resources_server.name=math_with_judge

# Test with weather resources
ng_run "+config_paths=[base.yaml]" \
Expand Down Expand Up @@ -218,7 +218,7 @@ omegaconf.errors.MissingMandatoryValue: Missing mandatory value: policy_api_key
### Problem: Invalid Server References
```
AssertionError: Could not find type='resources_servers' name='typo_weather'
in the list of available servers: [simple_weather, library_judge_math, ...]
in the list of available servers: [simple_weather, math_with_judge, ...]
```
**Fix**: Check your server name spelling and ensure the config is loaded.

Expand Down Expand Up @@ -312,7 +312,7 @@ policy_model:
# env.yaml - organize related configs
math_training_config_paths:
- responses_api_models/openai_model/configs/openai_model.yaml
- resources_servers/library_judge_math/configs/library_judge_math.yaml
- resources_servers/math_with_judge/configs/math_with_judge.yaml
- responses_api_agents/simple_agent/configs/simple_agent.yaml

weather_demo_config_paths:
Expand Down
18 changes: 9 additions & 9 deletions docs/about/concepts/rollout-collection-fundamentals.md
Original file line number Diff line number Diff line change
Expand Up @@ -85,18 +85,18 @@ Each line in your input JSONL file should follow this schema:

## Hands-On: Generating Your First Rollouts

Let's generate rollouts using the **MultiNeedle** resource server, which tests reading comprehension across long documents.
Let's generate rollouts using the **Example Multi Step** resource server, which tests reading comprehension across long documents.

### Step 1: Start the MultiNeedle Agent
### Step 1: Start the Example Multi Step Agent

```bash
# Start the multineedle agent server
# Start the example_multi_step agent server
config_paths="responses_api_models/openai_model/configs/openai_model.yaml,\
resources_servers/multineedle/configs/multineedle.yaml"
resources_servers/example_multi_step/configs/example_multi_step.yaml"
ng_run "+config_paths=[${config_paths}]"
```

**✅ Success Check**: You should see 3 servers running including the `multineedle_simple_agent`.
**✅ Success Check**: You should see 3 servers running including the `example_multi_step_simple_agent`.

### Step 3: Generate Rollouts

Expand All @@ -105,9 +105,9 @@ ng_run "+config_paths=[${config_paths}]"
In a separate terminal, run:
```bash
# Generate rollouts from the dataset
ng_collect_rollouts +agent_name=multineedle_simple_agent \
+input_jsonl_fpath=resources_servers/multineedle/data/example.jsonl \
+output_jsonl_fpath=results/multineedle_rollouts.jsonl \
ng_collect_rollouts +agent_name=example_multi_step_simple_agent \
+input_jsonl_fpath=resources_servers/example_multi_step/data/example.jsonl \
+output_jsonl_fpath=results/example_multi_step_rollouts.jsonl \
+limit=5 \
+num_repeats=2 \
+num_samples_in_parallel=3 \
Expand All @@ -124,7 +124,7 @@ ng_collect_rollouts +agent_name=multineedle_simple_agent \

```bash
# Launch the rollout viewer
ng_viewer +jsonl_fpath=results/multineedle_rollouts.jsonl
ng_viewer +jsonl_fpath=results/example_multi_step_rollouts.jsonl
```

**What you'll see**: An interactive viewer showing agent reasoning, tool calls, and verification scores for each rollout.
Expand Down
Loading