Nemotron 3.5 Super VL 120B A12B Pricing, Benchmarks & Channels
Nemotron 3.5 Super VL 120B A12B is a 2026-10-15 model from Thinking Machines featuring a 262K context window and extended reasoning capabilities. Lowest rate on Together AI ($0.15 in / $0.5 out per 1M tokens).
Input Price / 1M
$0.57
Output Price / 1M
$1.44
Cache Read / 1M
$0.114
Context Window
262K
Generation Speed
118 TPS
Access Grade
A · Global
F.A.R.E. Four-Dimension Score Breakdown
Scenario Capability Breakdown (0–100):
💻 Coding
72/100
✍️ Writing
58/100
📚 Long-Doc RAG
67/100
👁️ Vision OCR
70/100
F.A.R.E. CAPABILITY RADAR
Nemotron 3.5 Super VL 120B A12B Multi-Platform Price Comparison (2 Verified Channels)
All prices are in USD per 1M tokens. Domestic developers can prioritize channels marked "Direct in CN" for compliant access and local billing.
| Platform / Gateway | Network Access | Input ($/1M) | Output ($/1M) | Cache Hit ($/1M) | Billing & Notes | Link |
|---|---|---|---|---|---|---|
| Together AI Lowest Price | Global Network | $0.15 | $0.5 | $0.02 | Global cloud billing | Visit Official → |
| Thinking Machines Official | Global Network | $0.57 | $1.44 | $0.114 | Global cloud billing | Visit Official → |
✅ Core Strengths & Ideal Workloads
- Open weights available for private self-hosting (~74GB VRAM at INT4)
- Supports 262K context window and prompt caching ($0.114/M read)
⚠️ Limitations & Practical Caveats
- Frontier reasoning & agentic models (Inkling & Inkling 2 Small) available across OpenRouter, Together, and Nebius.
- Extended thinking generates internal reasoning tokens (~1.8x output multiplier)