Instructions to use nitrai-research/Moderato-V1-Pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nitrai-research/Moderato-V1-Pro with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="nitrai-research/Moderato-V1-Pro")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("nitrai-research/Moderato-V1-Pro", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nitrai-research/Moderato-V1-Pro with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nitrai-research/Moderato-V1-Pro"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nitrai-research/Moderato-V1-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nitrai-research/Moderato-V1-Pro

SGLang

How to use nitrai-research/Moderato-V1-Pro with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "nitrai-research/Moderato-V1-Pro" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nitrai-research/Moderato-V1-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "nitrai-research/Moderato-V1-Pro" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nitrai-research/Moderato-V1-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nitrai-research/Moderato-V1-Pro with Docker Model Runner:
```
docker model run hf.co/nitrai-research/Moderato-V1-Pro
```

Add formal evaluation results for benchmark leaderboards

by NitrAI - opened Aug 27

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+69

-70

YAML Metadata Error:Invalid content in Eval Result file .eval_results/results.yaml

Check out the documentation for more information.

Show details

Task ID "hle" does not match any task in dataset "cais/hle". Available: none

Files changed (8) hide show

.eval_results/deep-swe.yaml +0 -7
.eval_results/extractbench.yaml +0 -28
.eval_results/gpqa.yaml +0 -7
.eval_results/hle.yaml +0 -7
.eval_results/results.yaml +69 -0
.eval_results/swe-bench_pro.yaml +0 -7
.eval_results/terminal-bench-2.1.yaml +0 -7
.eval_results/wildclawbench.yaml +0 -7

.eval_results/deep-swe.yaml DELETED Viewed

@@ -1,7 +0,0 @@
-- dataset:
-    id: datacurve/deep-swe
-    task_id: deep_swe
-  value: 53.2
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card

.eval_results/extractbench.yaml DELETED Viewed

@@ -1,28 +0,0 @@
-- dataset:
-    id: llamaindex/ExtractBench
-    task_id: mean
-  value: 88.65
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card
-- dataset:
-    id: llamaindex/ExtractBench
-    task_id: short
-  value: 94.2
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card
-- dataset:
-    id: llamaindex/ExtractBench
-    task_id: medium
-  value: 86.4
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card
-- dataset:
-    id: llamaindex/ExtractBench
-    task_id: long
-  value: 36.8
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card

.eval_results/gpqa.yaml DELETED Viewed

@@ -1,7 +0,0 @@
-- dataset:
-    id: Idavidrein/gpqa
-    task_id: diamond
-  value: 90
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card

.eval_results/hle.yaml DELETED Viewed

@@ -1,7 +0,0 @@
-- dataset:
-    id: cais/hle
-    task_id: hle
-  value: 38.4
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card

.eval_results/results.yaml ADDED Viewed

	@@ -0,0 +1,69 @@

+- dataset:
+    id: cais/hle
+    task_id: hle
+  value: 38.40
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "no-tools"
+- dataset:
+    id: Idavidrein/gpqa
+    task_id: diamond
+  value: 90.00
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "Diamond split, 0-shot CoT"
+- dataset:
+    id: ScaleAI/SWE-bench_Pro
+    task_id: SWE_Bench_Pro
+  value: 63.30
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "Pass@1"
+- dataset:
+    id: datacurve/deep-swe
+    task_id: deep_swe
+  value: 53.20
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "Pass@1 (v1.1)"
+- dataset:
+    id: harborframework/terminal-bench-2.1
+    task_id: terminalbench_2_1
+  value: 79.50
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "Pass@1 (Terminus)"
+- dataset:
+    id: internlm/WildClawBench
+    task_id: overall
+  value: 52.20
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "Pass@1 (Overall)"
+- dataset:
+    id: llamaindex/ExtractBench
+    task_id: mean
+  value: 88.65
+  date: "2026-08-26"
+  source:
+    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
+    name: "Moderato-V1-Pro Technical Report"
+  notes: "Mean score"

.eval_results/swe-bench_pro.yaml DELETED Viewed

@@ -1,7 +0,0 @@
-- dataset:
-    id: ScaleAI/SWE-bench_Pro
-    task_id: SWE_Bench_Pro
-  value: 63.3
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card

.eval_results/terminal-bench-2.1.yaml DELETED Viewed

@@ -1,7 +0,0 @@
-- dataset:
-    id: harborframework/terminal-bench-2.1
-    task_id: terminalbench_2_1
-  value: 79.5
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card

.eval_results/wildclawbench.yaml DELETED Viewed

@@ -1,7 +0,0 @@
-- dataset:
-    id: internlm/WildClawBench
-    task_id: overall
-  value: 52.2
-  source:
-    url: https://huggingface.co/nitrai-research/Moderato-V1-Pro
-    name: Model Card