Compare · ModelsLive · 2 picked · head to head
Gemini 4 Carbon vs Kimi K3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 4 Carbon wins on 13/13 benchmarks
Gemini 4 Carbon wins 13 of 13 shared benchmarks. Leads in agentic · reasoning · knowledge.
Category leads
agentic·Gemini 4 Carbonreasoning·Gemini 4 Carbonknowledge·Gemini 4 Carbonmath·Gemini 4 Carboncoding·Gemini 4 Carbongeneral·Gemini 4 Carbon
Head to head · from the data · data as of 2026-11-17
- Benchmarks · Gemini 4 Carbon leads on 13 of 13 shared benchmarks (APEX-Agents, ARC-AGI, ARC-AGI-2, Chess Puzzles, 9 more), Kimi K3 on 0
- Price · Gemini 4 Carbon input $2.00/M vs Kimi K3 $0.19/M, Kimi K3 is 90% cheaper (OpenRouter list price, as of 2026-11-17)
- Cheapest provider · Gemini 4 Carbon at Google AI Studio $1.00/M input (of 7 offers) as of 2026-11-17 · Kimi K3 at Wafer $0.15/M input (of 27 offers) as of 2026-11-17
- Context · Gemini 4 Carbon and Kimi K3 both 1.0M tokens
- Released · Gemini 4 Carbon Nov 10, 2026 vs Kimi K3 Jul 16, 2026 · 4 months apart
- Price history · Gemini 4 Carbon OpenRouter input list price had no move of 5% or more in the 90 days to 2026-11-17
Sources · scores from public leaderboards, list prices and provider offers from OpenRouter, price moves from the BenchGecko changelog (5% or more, held in the next daily check), Gecko tests run by BenchGecko.
Hype vs Reality
Attention vs performance
Gemini 4 Carbon
#1 by perf·#2 by attention
Kimi K3
#70 by perf·#20 by attention
Best value
Gemini 4 Carbon
2.0x better value than Kimi K3
Gemini 4 Carbon
14.3 pts/$
$6.00/M
Kimi K3
7.1 pts/$
$8.10/M
Vendor risk
Who is behind the model
Google DeepMind
$3.49T·Tier 1
Moonshot AI
$50.0B·Tier 1
Head to head
13 benchmarks · 2 models
Gemini 4 CarbonKimi K3
APEX-Agents
Gemini 4 Carbon leads by +36.4
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Gemini 4 Carbon
84.2
Kimi K3
47.8
ARC-AGI
Gemini 4 Carbon leads by +4.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Gemini 4 Carbon
98.5
Kimi K3
94.5
ARC-AGI-2
Gemini 4 Carbon leads by +33.8
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Gemini 4 Carbon
94.2
Kimi K3
60.4
Chess Puzzles
Gemini 4 Carbon leads by +43.1
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 4 Carbon
76.8
Kimi K3
33.7
FrontierMath-Tier-4-v2-Private
Gemini 4 Carbon leads by +56.1
Gemini 4 Carbon
95.1
Kimi K3
39.0
FrontierMath-Tiers-1-3-v2-Private
Gemini 4 Carbon leads by +19.3
Gemini 4 Carbon
95.3
Kimi K3
76.0
Frontierswe
Gemini 4 Carbon leads by +35.3
Gemini 4 Carbon
61.2
Kimi K3
25.9
Furniture Assembly
Gemini 4 Carbon leads by +61.9
Gemini 4 Carbon
67.9
Kimi K3
6.0
GPQA diamond
Gemini 4 Carbon leads by +4.7
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemini 4 Carbon
99.3
Kimi K3
94.6
Mystery Game Puzzles
Gemini 4 Carbon leads by +51.8
Gemini 4 Carbon
70.3
Kimi K3
18.5
OTIS Mock AIME 2024-2025
Gemini 4 Carbon leads by +2.8
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 4 Carbon
100.0
Kimi K3
97.2
SimpleBench
Gemini 4 Carbon leads by +38.3
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Gemini 4 Carbon
91.1
Kimi K3
52.8
SimpleQA Verified
Gemini 4 Carbon leads by +27.1
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 4 Carbon
77.7
Kimi K3
50.6
Full benchmark table
| Benchmark | Gemini 4 Carbon | Kimi K3 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 84.2 | 47.8 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 98.5 | 94.5 |
ARC-AGI-2 ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data. | 94.2 | 60.4 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 76.8 | 33.7 |
FrontierMath-Tier-4-v2-Private | 95.1 | 39.0 |
FrontierMath-Tiers-1-3-v2-Private | 95.3 | 76.0 |
Frontierswe | 61.2 | 25.9 |
Furniture Assembly | 67.9 | 6.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 99.3 | 94.6 |
Mystery Game Puzzles | 70.3 | 18.5 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 100.0 | 97.2 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 91.1 | 52.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 77.7 | 50.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $10.00 | 1.0M tokens (~524 books) | $40.00 | |
| $0.19 | $16.00 | 1.0M tokens (~524 books) | $41.44 |
Gemini 4 Carbon vs Kimi K3 · questions answered from the data
Gemini 4 Carbon leads on 13 of 13 shared benchmarks (APEX-Agents, ARC-AGI, ARC-AGI-2, Chess Puzzles, 9 more), Kimi K3 on 0