Inference
EN
NVIDIA

Nemotron 3.5 Bolt E4B

Nemotron 3.5 Bolt E4B is a reasoning model from NVIDIA. It has 3 scored results in 3 categories. It has no index.

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-10
UnrankedEnough results for an index.
IndexAll categories, out of 100.UnrankedMedian 51.0
CoverageIndex weight that has results.40%
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.64KMedian 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.1 counted37.5—
CodingCode writing and repair.None counted——
ReasoningLogic problems and puzzles.1 counted38.1—
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.None counted——
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted46.1—
MathMath problems.None counted——

Results

3 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Artificial Analysis IFBenchInstruction61.1%62 / 146
τ²-Bench Tool-Agent-User EvaluationAgentic52.3%94 / 149
LongBench v2Reasoning48.7%22 / 25
RULER at a 64K-token contextReasoning85.8%—
Brevity Benchmark (NVIDIA)Instruction82.8%—
IHEval: instruction hierarchy evaluationInstruction82.4%—
StructEval text-format subsetInstruction80.0%—
RoleMRCInstruction79.1%—
RAGTruth: provider-reported F1Knowledge77.9%—
ToolSandboxAgentic74.1%—
RoleBench (RoleLLM)Instruction71.2%—
LongMemEvalReasoning69.9%—
RelyToolBenchAgentic69.3%—
Berkeley Function Calling Leaderboard v3Agentic67.3%—
SysBench: system message followingInstruction67.2%—
When2Call Multi-Turn (NVIDIA)Agentic60.5%—
MultiChallenge: provider-reported runsInstruction55.6%—
InverseIFEvalInstruction45.3%—
GaRAGeKnowledge40.7%—
EQ-Bench 3Instruction38.1%—
AgentIF: instruction following in agentic scenariosInstruction22.9%—

Sources

BenchLMCC BY-NC 4.0

Questions

What is the context window of Nemotron 3.5 Bolt E4B?
64K tokens.
Is Nemotron 3.5 Bolt E4B open source?
Yes. Its weights are published.
When was Nemotron 3.5 Bolt E4B released?
October 2026.

More from NVIDIA

All models
Nemotron 3.5 Bolt E4B BF16UnrankedNemotron 3.5 Super VLUnrankedNemotron 3.5 Super VL 120B A12B51.0Nemotron 3.5 Super VL 120B A12B NVFP450.8Nemotron 3.5 Bolt 74B-A9B46.4