Inference
EN
NVIDIA

Nemotron 3.5 Bolt 74B-A9B

Nemotron 3.5 Bolt 74B-A9B is a reasoning model from NVIDIA. 11 benchmarks count toward its index, in 4 categories.

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-10
RankedEnough results for an index.
IndexAll categories, out of 100.46.4±7.9Median 51.0
CoverageIndex weight that has results.70%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.256KMedian 262K
EloLMArena rating and rank.—Median 1422

Run locally

74.04B parameters
16-bit
156.5 GB
8-bit
83.6 GB
4-bit
48.1 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy fits in 64 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.6 counted42.3171 / 218
CodingCode writing and repair.1 counted47.6—
ReasoningLogic problems and puzzles.2 counted50.1115 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.2 counted53.0112 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.None counted——

Results

11 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Critical Physics TasksReasoning4.3%91 / 214
Artificial Analysis Humanity's Last ExamKnowledge24.7%100 / 222
PinchBenchAgentic87.8%5 / 10
Artificial Analysis Long Context ReasoningReasoning69.7%113 / 217
GDPval-AA normalizedAgentic24.0%77 / 133
GPQA Diamond · BenchLMKnowledge86.4%51 / 84
BrowseCompAgentic66.8%46 / 63
Artificial Analysis SciCodeCoding39.7%89 / 114
MCP AtlasAgentic52.2%44 / 52
Terminal-Bench 2.1 · BenchLMAgentic51.0%74 / 86
Humanity's Last Exam without toolsKnowledge25.3%46 / 53
τ³-Bench Tool-Agent-User EvaluationAgentic17.0%28 / 32
WebArena Web Agent BenchmarkAgentic50.9%—

Sources

BenchLMCC BY-NC 4.0

History

5 releases
Index45.8 → 46.4
23 Oct 202612 Nov 2026

Questions

How good is Nemotron 3.5 Bolt 74B-A9B?
Its index is 46.4 out of 100. That is place 174 of 283 ranked models.
What is Nemotron 3.5 Bolt 74B-A9B best at?
Knowledge: 53.0 out of 100, place 112 of 268.
What is the context window of Nemotron 3.5 Bolt 74B-A9B?
256K tokens.
Is Nemotron 3.5 Bolt 74B-A9B open source?
Yes. Its weights are published.
When was Nemotron 3.5 Bolt 74B-A9B released?
October 2026.

Badge

Nemotron 3.5 Bolt 74B-A9B on Inference360: place 174 of 283, index 46.4
HTML
<a href="https://inference360.ai/models/nemotron-3-5-bolt-74b-a9b"><img src="https://inference360.ai/badges/nemotron-3-5-bolt-74b-a9b.svg" alt="Nemotron 3.5 Bolt 74B-A9B on Inference360: place 174 of 283, index 46.4" height="20"></a>
Markdown
[![Nemotron 3.5 Bolt 74B-A9B on Inference360: place 174 of 283, index 46.4](https://inference360.ai/badges/nemotron-3-5-bolt-74b-a9b.svg)](https://inference360.ai/models/nemotron-3-5-bolt-74b-a9b)

Compare with

Mercury 2.546.2Gemini 2.5 Pro46.6Nemotron 3.5 Super VL 120B A12B NVFP450.8

More from NVIDIA

All models
Nemotron 3.5 Super VL 120B A12B51.0Nemotron 3.5 Super VL 120B A12B NVFP450.8Nemotron 3.5 Bolt 74B-A9B BF16UnrankedNemotron 3.5 Super VLUnrankedNemotron 3.5 Bolt E4BUnranked