Today's price · per 1M tokens
Input
$1.20
per 1M tokens
Output
$1.20
per 1M tokens
Blended
$1.20
blended $/1M — 3:1 weighted input:output
Overview
NVIDIA's Llama-3.1-based Nemotron 70B Instruct, a 70B open-weights model tuned with RLHF for instruction-following. Last sold as a hosted API by DeepInfra at $1.20/$1.20 per 1M tokens, delisted 2026-07-16; NVIDIA itself never sold per-token access and its free build.nvidia.com endpoint is deprecated. Open weights remain available for self-hosting.
Deploy this open model on rented GPUs
Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.
Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.
Capabilities
struck through = not supportedBenchmark performance
accuracy % · higher is betterWe hold no per-benchmark accuracy scores for this model yet, so it has no accuracy average. That is a gap in our coverage, not a sign the model is untested — independent evaluators often publish a composite index for a new model long before releasing its per-benchmark numbers. The percentile above is its standing across the independent composites below.
- HF Open LLM v2 avg: 36.9 (0–100)
Source: HuggingFace Open LLM Leaderboard · updated Jul 1, 2026 · See full rankings →
Specifications
- Provider
- DeepInfra
- Context window
- 128K tokens
- Modality
- text
- Parameters
- 70B
- Open source
- Yes — open weights available
- Released
- Jun 1, 2025
- Status
- Retired ended Jul 16, 2026
- Last updated
- Aug 8, 2026
- Tags
Availability verified: Aug 9, 2026 — per DeepInfra's own deprecation notice