Doc updates

Signed-off-by: Igor Gitman <igitman@nvidia.com>

Author: Kipok

README.md

@@ -9,11 +9,11 @@ Here are some of the things we support.
 - [Model evaluation](https://nvidia.github.io/NeMo-Skills/pipelines/evaluation): Evaluate your models on many popular benchmarks
     - Math problem solving: math, aime24, aime25, omni-math (and many more)
     - Formal proofs in Lean: minif2f, proofnet
-    - Coding skills: lcb, human-eval, mbpp
+    - Coding skills: livecodebench, human-eval, mbpp
     - Chat/instruction following: ifeval, arena-hard, mt-bench
     - General knowledge: mmlu, mmlu-pro, gpqa
     - Long context: ruler
-- [Model training](https://nvidia.github.io/NeMo-Skills/pipelines/training): Train models at speed-of-light using [NeMo-Aligner](https://github.com/NVIDIA/NeMo-Aligner/).
+- [Model training](https://nvidia.github.io/NeMo-Skills/pipelines/training): Train models at speed-of-light using [NeMo-Aligner](https://github.com/NVIDIA/NeMo-Aligner/), [NeMo-RL](https://github.com/NVIDIA/NeMo-RL/) or [verl](https://github.com/volcengine/verl).
 
 You can find the full documentation [here](https://nvidia.github.io/NeMo-Skills/).
 To get started, follow this [tutorial](https://nvidia.github.io/NeMo-Skills/basics),

docs/basics/index.md

@@ -98,9 +98,9 @@ config might look like
 executor: local
 
 containers:
-  trtllm: igitman/nemo-skills-trtllm:0.5.0
-  vllm: igitman/nemo-skills-vllm:0.5.3
-  nemo: igitman/nemo-skills-nemo:0.5.3
+  trtllm: igitman/nemo-skills-trtllm:0.6.1
+  vllm: vllm/vllm-openai:v0.9.0
+  nemo: igitman/nemo-skills-nemo:0.6.1
   # ... there are some more containers defined here
 
 env_vars:

docs/index.md

@@ -13,11 +13,12 @@ Here are some of the things we support.
 - [Model evaluation](pipelines/evaluation.md): Evaluate your models on many popular benchmarks
     - Math problem solving: math, aime24, aime25, omni-math (and many more)
     - Formal proofs in Lean: minif2f, proofnet
-    - Coding skills: lcb, human-eval, mbpp
+    - Coding skills: livecodebench, human-eval, mbpp
     - Chat/instruction following: ifeval, arena-hard, mt-bench
     - General knowledge: mmlu, mmlu-pro, gpqa
     - Long context: ruler
-- [Model training](pipelines/training.md): Train models at speed-of-light using [NeMo-Aligner](https://github.com/NVIDIA/NeMo-Aligner/).
+- [Model training](pipelines/training.md): Train models at speed-of-light using [NeMo-Aligner](https://github.com/NVIDIA/NeMo-Aligner/), [NeMo-RL](https://github.com/NVIDIA/NeMo-RL/) or [verl](https://github.com/volcengine/verl).
+
 
 To get started, follow this [tutorial](basics/index.md), browse available [pipelines](./pipelines/index.md) or run `ns --help` to see all available
 commands and their options.

docs/pipelines/evaluation.md

@@ -11,7 +11,7 @@ We support many popular benchmarks and it's easy to add new in the future. E.g.
 
 - Math problem solving: math, aime24, aime25, omni-math (and many more)
 - Formal proofs in Lean: minif2f, proofnet
-- Coding skills: lcb, human-eval, mbpp
+- Coding skills: livecodebench, human-eval, mbpp
 - Chat/instruction following: ifeval, arena-hard, mt-bench
 - General knowledge: mmlu, mmlu-pro, gpqa
 - Long context: ruler

docs/pipelines/generation.md

@@ -320,12 +320,13 @@ Here are some suggestions on how to make your generation jobs more efficient
    random seed separately. We will run `.format(random_seed=random_seed)` on your command which lets you run the same
    logic on each output file, e.g.
 
-        ```python
-        generate(
-            # ...
-            postprocess_cmd="python /nemo_run/code/my_script.py --input <output_dir>/output-{random_seed}.jsonl"
-        )
-        ```
-
-    If you need to run some logic that aggregates information from across all random seeds, you can instead schedule
-    a dependent [run_cmd command](./run-cmd.md).
+    ```python
+    cmd = f"python /nemo_run/code/my_script.py {output_dir}/output.jsonl"
+    generate(
+        # ...
+        postprocess_cmd=cmd
+    )
+    ```
+
+   If you need to run some logic that aggregates information from across all random seeds, you can instead schedule
+   a dependent [run_cmd command](./run-cmd.md).