The Model Index
Up to four models

Nemotron 3 Ultra vs Gemini 3.5 Flash-Lite

On the index (GPT-4 = 100), Nemotron 3 Ultra scores 141 and Gemini 3.5 Flash-Lite 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Nemotron 3 Ultra
Presets
Nemotron 3 Ultra
141124–157
WeightsOpen
ReleasedJun 2026
Price in / out$0.50 / $2.2
Context262K
Core tests taken3 of 9
Gemini 3.5 Flash-Lite
141125–156
WeightsClosed
ReleasedJul 2026
Price in / out$0.30 / $2.5
Context1.05M
Core tests taken5 of 9

Nemotron 3 Ultra and Gemini 3.5 Flash-Lite are too close to call (0 points apart, within the uncertainty), ahead on 1 of 3 shared core tests, at a similar price.

Test by testthird-party results

Nemotron 3 UltraGemini 3.5 Flash-Lite
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)GPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)CODINGCode MigrationCode MigrationLiveCodeBenchLiveCodeBenchProgramBenchProgramBenchSWE-benchSWE-benchVibe Code Bench v1.1Vibe Code Bench v1.1WeirdMLWeirdMLAGENTSAPEX-AgentsAPEX-AgentsDTBenchDTBenchLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)CorpFin v2CorpFin v2EMBEMBLegalBenchLegalBenchLegal Research BenchLegal Research BenchMedCodeMedCodeTaxEval v2TaxEval v2
Every result as a table
TestNemotron 3 UltraGemini 3.5 Flash-Lite
Reasoning & knowledge
GPQA Diamond (Epoch AI)85.4%83.3%
GPQA Diamond (Vals AI)86.1%83.8%
MMLU Pro85.8%85.8%
Maths
AIME-style maths (OTIS mock)86.7%71.1%
Coding
Code Migration4.9%6.2%
LiveCodeBench86.0%79.0%
ProgramBench0.0%0.0%
SWE-bench69.0%75.0%
Vibe Code Bench v1.17.6%37.2%
WeirdML43.5%39.0%
Agents
APEX-Agents22.7%29.3%
DTBench90.1%83.5%
LMCA36.9%37.6%
Mystery game puzzles20.0%19.0%
Professional work
Finance Agent (v2)37.7%47.4%
CorpFin v265.5%60.7%
EMB31.7%43.2%
LegalBench82.1%84.1%
Legal Research Bench15.4%13.9%
MedCode38.6%43.5%
TaxEval v273.1%72.6%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG