MDL
Llama Nemotron Super 49B v1.5
NVIDIA
Verified against Artificial Analysis · Jul 13, 2026
Capability
65
percentile index
Price /M
$0.4 / $0.4
in / out
Context
—
Avg score
50.1
11 benchmarks
AA Intelligence
12.4
index
Speed
51 tok/s
TTFT 0.2s
Released
Jul 25, 2025
About
Llama Nemotron Super 49B v1.5 is a open-weights model in the Llama family from NVIDIA. Benchmarked on 11 evals, averaging 50.1.
Benchmark scores · 11
avg 50.1 — higher is betterThe research behind this model
all papers mentioning it →Llama-Nemotron: Efficient Reasoning Models1 cites · May 2, 2025Benchmarking Patent Embeddings: A Multi-Task Evaluation of 22 Models Across Retrieval, Classification, and ClusteringMay 22, 2026Quantization-Aware Distillation for NVFP4 Inference Accuracy RecoveryJan 27, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed51 tok/s
Latency (TTFT)0.2s
First answer40s
Price in / out$0.4 / $0.4 per 1M
Blended price$0.4 / 1M
Math index76.7
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$0.4/M