GetInfoUs
Head-to-Head Model Comparison

Gemini 3.5 Flash-Lite vs Grok 4.6

Benchmarks, API Token Pricing, Context Window & Speed Comparison

Geprüft vom GetInfoUs-ForschungsteamZuletzt geprüft: October 2026Redaktionelle Richtlinien
Teilen
Related topics (12)
AI CompareAI BenchmarkToken PricingModel CostModel SpecsSWE-bench CodingModel VersusWhich AI ModelLLM Speed & LatencyReasoning ModelsExtended ThinkingPrompt Caching ROI
Gemini 3.5 Flash-Lite vs Grok 4.6: Benchmarks, Pricing & Speed Comparison

Gemini 3.5 Flash-Lite von Google und Grok 4.6 von xAI gehören zu den führenden KI-Modellen. Gemini 3.5 Flash-Lite bietet ein Kontextfenster von 1.05M Tokens bei API-Preisen von $0.40/$2.00 pro 1M Tokens, während Grok 4.6 500K Tokens Kontext zu $2.00/$6.00 pro 1M Tokens bereitstellt. Hier finden Sie den vollständigen Benchmark-, Latenz- und Kostenvergleich.

Direktvergleich (VS)

Beliebige Modelle wechseln & vergleichen

Select any AI models from our verified benchmark database to generate live side-by-side matrices.

Profile
Profile
Multi-Model Arena Stack:
5-Model Matrix Workbench
Popular Matchups:
Google

Gemini 3.5 Flash-Lite

Ultraschnelle Echtzeit-Inferenz

Gemini 3.5 Flash-Lite von Google ist ein KI-Modell für ultraschnelle Echtzeit-Antworten, niedrige Latenz und Hochdurchsatz-Routing mit 1.0M Kontextfenster, 240 Token/s Durchsatz, einem Intelligenz-Index von 69/100 und $0.4/$2 pro 1M Token.

Intelligence Index69 / 100
Input Token:

$0.40 / 1M

Output Token:

$2.00 / 1M

Context Window:

1.05M tokens

SWE-bench Verified:

N/A

xAI

Grok 4.6

Frontier-Flaggschiff-Modell

Grok 4.6 von xAI ist ein KI-Modell für anspruchsvolles Enterprise-Reasoning, Softwareentwicklung und produktive KI-Workflows mit 500K Kontextfenster, 115 Token/s Durchsatz, einem Intelligenz-Index von 77/100 und $2/$6 pro 1M Token.

Intelligence Index77 / 100
Input Token:

$2.00 / 1M

Output Token:

$6.00 / 1M

Context Window:

500K tokens

SWE-bench Verified:

75.8%

Wichtigste Unterschiede auf einen Blick

  • Gemini 3.5 Flash-Lite is 3.6x cheaper for API usage (input + output tokens combined).
  • Gemini 3.5 Flash-Lite has a larger context window (1M vs 500K tokens).
  • Gemini 3.5 Flash-Lite is faster at 240 tok/s vs 115 tok/s.
Need definitions for token pricing, context, or benchmarks?Explore AI Glossary →
FeatureGemini 3.5 Flash-LiteGrok 4.6
Intelligence Index
Composite reasoning & knowledge score (0-100)
6977
Provider
GooglexAI
Input Price
Cost per million input tokens
$0.40 / 1M$2.00 / 1M
Output Price
Cost per million output tokens
$2.00 / 1M$6.00 / 1M
Blended Price
Weighted 7:2:1 cache:input:output per 1M
$0.35 / 1M$1.35 / 1M
Context Window
Maximum input + output tokens
1.05M tokens~1,573 A4 pages of text500K tokens~750 A4 pages of text
Max Output Tokens
Per single request
33K66K
Throughput Speed
Tokens generated per second
240 tok/s115 tok/s
SWE-bench Verified
Real-world software engineering
N/A75.8%
GPQA Diamond
Graduate-level reasoning
N/A77.2%
Cached Input Price
Prompt caching discount per 1M tokens
$0.10 / 1M$0.50 / 1M

Wer gewinnt in welcher Disziplin?

At-a-glance decision framework
Coding

Evaluations ongoing

Cost
Gemini 3.5 Flash-Lite
3.6x cheaper monthly runtime
Speed
Gemini 3.5 Flash-Lite
+109% faster generation
Context
Gemini 3.5 Flash-Lite
1.05M tokens window

Geprüfte Benchmark-Ergebnisse im Vergleich

Official evaluation metrics verified by independent third-party research labs.

Gemini 3.5 Flash-LiteGrok 4.6
SWE-bench Verified (Autonomous Coding)

Industry benchmark evaluating real-world GitHub issue resolution and multi-file debugging.

N/Avs75.8%
GPQA Diamond (Doctorate-Level Science & Math)

Challenging questions created by domain experts to test complex PhD-level reasoning.

N/Avs77.2%
MMLU-Pro (Multi-task Language Understanding)

Hardened reasoning and academic knowledge benchmark with reduced lucky-guess probability.

81.2%vs88%
LMSYS Chatbot Arena Elo

Crowdsourced human preference evaluation based on blind, head-to-head model battles.

1375ptsvs1450pts
Humanity's Last Exam (Frontier Expert Knowledge)

Extremely difficult, closed-ended questions spanning the frontier of human expert knowledge.

N/Avs14.8%
Terminal-Bench (Agentic Coding & Terminal Use)

Autonomous multi-step tasks executed inside a real command-line terminal environment.

N/Avs51%
SciCode (Scientific Computing)

Research-grade coding problems drawn from real scientific and numerical workloads.

N/Avs49.2%
LiveCodeBench (Competitive Coding)

Contamination-free competitive programming problems collected continuously over time.

N/Avs75%
AIME (Advanced Math Competition)

American Invitational Mathematics Examination — multi-step Olympiad-level reasoning.

N/Avs89.5%
Long-Context Reasoning

Accuracy when reasoning over very large inputs that fill the model context window.

N/Avs89%
Generation Speed (Throughput)

Tokens emitted per second during active generation streaming.

240 tok/s vs115 tok/s

Latenz & Antwortzeit

Time to first token (TTFT) and estimated end-to-end response time for a typical 500-token response.

Time to First Token (TTFT)Lower is better
Gemini 3.5 Flash-Lite0.29s
Grok 4.60.58s
Est. End-to-End Response (500 tokens)TTFT + generation time
Gemini 3.5 Flash-Lite2.37s
Grok 4.64.93s

Interaktiver Token-Kosten-Simulator

Adjust sliders to calculate projected monthly cloud API spending with prompt caching discounts.

100K / day
10K500K1M
30K / day
5K150K300K
50% cached
0% (Cold)50%90% (Warm)
Gemini 3.5 Flash-Lite Monthly API Bill
$2.55 / month
$0.4/1M in · $2/1M out
Grok 4.6 Monthly API Bill
$9.15 / month
$2/1M in · $6/1M out
Gemini 3.5 Flash-Lite saves approximately $6.60 / month (72%) under this workload.3.6x Savings

Gemini 3.5 Flash-Lite · Strengths & Key Advantages

  • 1.0M Kontextfenster und bis zu 33K maximale Output-Token (MMLU-Pro 81.2% · Arena Elo 1375)
  • Hohe Generierungsgeschwindigkeit von 240 Token/Sekunde bei nur 290 ms Erstlatenz (TTFT)
  • Effiziente API-Struktur ($0.4/$2 pro 1M Token) inklusive Prompt-Caching, Tool-Calling und strikter JSON-Ausgabe
Considerations & Trade-offs
  • Proprietäres Cloud-Modell ohne Möglichkeit zum lokalen Download der Modellgewichte
  • Bei extrem komplexen mehrstufigen Forschungsfragen bieten schwere Reasoning-Spezialmodelle noch tiefere Analysen

Grok 4.6 · Strengths & Key Advantages

  • 500K Kontextfenster und bis zu 66K maximale Output-Token (SWE-bench Verified 75.8% · GPQA Diamond 77.2%)
  • Hohe Generierungsgeschwindigkeit von 115 Token/Sekunde bei nur 580 ms Erstlatenz (TTFT)
  • Effiziente API-Struktur ($2/$6 pro 1M Token) inklusive Prompt-Caching, Tool-Calling und strikter JSON-Ausgabe
Considerations & Trade-offs
  • Proprietäres Cloud-Modell ohne Möglichkeit zum lokalen Download der Modellgewichte
  • Bei hohem Anfragevolumen empfiehlt sich der Einsatz von Prompt-Caching und Batch-API zur Kostenoptimierung

Token Use & Task Economics

Estimated cost and token consumption to complete one representative agentic task.

MetricGemini 3.5 Flash-LiteGrok 4.6
Cost per Task
Estimated USD to finish one task
$0.55$0.65
Output Tokens / Task
Tokens generated per task
18K45K
Reasoning Tokens / Task
Thinking tokens burned per task
7K22K

Technische Spezifikationen

Architecture, release timeline, capabilities, and modalities.

FeatureGemini 3.5 Flash-LiteGrok 4.6
Official API Model ID (SDK)models/gemini-3-5-flash-litegrok-4-6-latest
API ProtocolGoogle Gemini API (generateContent) + OpenAI-Compatible SDKxAI REST API (OpenAI-Compatible)
Release Date2026-072026-08
Knowledge CutoffMarch 2026April 2026
ArchitectureLightweight efficient multimodal transformer1.5T Colossus Sparse MoE
ParametersConfidential1.5T (MoE)
Enterprise Rate Limits (RPM / TPM)10,000 – 30,000 RPM (Tier 5 Enterprise) · 4,000,000 – 10,000,000 TPM5,000 – 15,000 RPM (Production Tier) · 800,000 – 2,000,000 TPM
Long-Context Tier (>128K/200K)Standard rate up to 200K tokens; 2x input/output tier above 200K tokensFlat rate ($2/1M in) across full 500K window
Prompt Cache Policy & TTLImplicit + Explicit Context Caching at $0.1/1M with configurable TTLCached input at $0.5/1M tokens
Computer Use / GUI AgentSupported (Screen coordinate grounding, browser & desktop GUI action loops)API / CLI Tool-Use & DOM Agent ready (no raw pixel desktop click loop)
MCP & Parallel Tool CallingNative MCP Client/Server ready · Parallel function calling · 100% Strict JSON SchemaNative MCP Client/Server ready · Parallel function calling · 100% Strict JSON Schema
Web Search Grounding & SandboxBuilt-in Google Search Grounding + Managed Python Code Execution SandboxReal-time Web & X (Twitter) Search Grounding + Code Execution
Official Fine-Tuning SupportVertex AI Supervised Adapter Tuning & Context DistillationAPI Prompt Engineering & RAG Distillation
Multimodal Token BillingText BPE (~4 chars/token) · Vision (~258–765 tokens per image tile)Text BPE (~4 chars/token) · Vision (~258–765 tokens per image tile)
Quantization & Min GPU VRAMCloud-Hosted Proprietary (FP8 / BF16 Tensor Core Cluster) · Zero Local VRAM Required (100% Serverless Cloud API)Cloud-Hosted Proprietary (FP8 / BF16 Tensor Core Cluster) · Zero Local VRAM Required (100% Serverless Cloud API)
Safety & Frontier ClassificationGoogle DeepMind Frontier Safety Framework VerifiedxAI Production Safety & Alignment Verified
Enterprise Compliance (ZDR/HIPAA/SOC2)SOC2 Type II · GDPR · Zero Data Retention (ZDR) & HIPAA BAA Available on Enterprise TierSOC2 Type II · GDPR · Zero Data Retention (ZDR) & HIPAA BAA Available on Enterprise Tier
Input Modalitiestext, visiontext, vision
Output Modalitiestexttext

Gemini 3.5 Flash-Lite vs Grok 4.6: Schnelle Antworten (FAQ)

Which is more intelligent, Gemini 3.5 Flash-Lite or Grok 4.6?

Grok 4.6 is more intelligent. It scores 77 compared with 69 on the composite Intelligence Index.

Which model is faster?

Gemini 3.5 Flash-Lite is faster, generating 240 tokens per second compared with 115 tokens per second.

Which model is cheaper to run?

Gemini 3.5 Flash-Lite is cheaper at $0.35 per 1M tokens blended, compared with $1.35 (7:2:1 cache:input:output ratio).

Which model has lower latency?

Gemini 3.5 Flash-Lite has lower latency, with a time to first token of 0.29s compared with 0.58s.

Which model has a larger context window?

Gemini 3.5 Flash-Lite has a larger context window, supporting 1.05M tokens compared with 500K tokens.

Fazit

Sowohl Gemini 3.5 Flash-Lite (Optimal für ultraschnelle Echtzeit-Antworten, niedrige Latenz und Hochdurchsatz-Routing, Enterprise-RAG-Systeme und skalierbare API-Pipelines.) als auch Grok 4.6 (Optimal für anspruchsvolles Enterprise-Reasoning, Softwareentwicklung und produktive KI-Workflows, Enterprise-RAG-Systeme und skalierbare API-Pipelines.) überzeugen im Produktionseinsatz: Grok 4.6 führt bei komplexem Reasoning und Coding, Gemini 3.5 Flash-Lite liefert den höheren Token-Durchsatz und Gemini 3.5 Flash-Lite ist die kosteneffizienteste Wahl für große API-Workloads.

Häufig gestellte Fragen (FAQ)

Welches Modell ist besser für Coding und Reasoning – Gemini 3.5 Flash-Lite oder Grok 4.6?+

Im Gesamt-Intelligenzindex und bei Software-Benchmarks liegt Grok 4.6 vorn. Prüfen Sie oben die SWE-bench- und GPQA-Werte für Ihren Anwendungsfall.

Wie unterscheiden sich die API-Preise von Gemini 3.5 Flash-Lite und Grok 4.6?+

Gemini 3.5 Flash-Lite kostet $0.40 / $2.00 pro 1M Input-/Output-Tokens, Grok 4.6 liegt bei $2.00 / $6.00. Beim gewichteten Gesamtpreis ist Gemini 3.5 Flash-Lite günstiger.

Welches Modell hat das größere Kontextfenster und die höhere Geschwindigkeit?+

Gemini 3.5 Flash-Lite bietet 1.05M Tokens Kontext gegenüber 500K Tokens bei Grok 4.6. Bei der Generierungsgeschwindigkeit liegt Gemini 3.5 Flash-Lite vorn.

AI Terminology & Guide
Explore Glossary →

Understand key metrics like context window, token pricing, reasoning mode, and SWE-bench scores in our comprehensive AI glossary.

Tokens & PricingContext WindowAI BenchmarksReasoning ModelsTool CallingModel Context Protocol (MCP)

Related reading

Benchmark scores, token prices, and rate limits evolve rapidly as AI labs release model updates. Always verify current specifications and pricing on the official provider documentation.

Offizielle Quellen: Offizielle API-Dokumentation von Google und xAI, Systemkarten und unabhängige Benchmark-Auswertungen (2026).