Gemini 4 Carbon vs Kimi K3.1 Pro
Full profiles: Gemini 4 Carbon (Google) and Kimi K3.1 Pro (Moonshot AI): every benchmark, price and rank.
Updated · Data from 7 sources
As of 29 Oct 2026, Gemini 4 Carbon and Kimi K3.1 Pro are neck and neck on quality (65.1 vs 61.3 on the BenchLeader Index, inside the ±7.2 margin), and Kimi K3.1 Pro is 2.0× cheaper at $4.00 vs $8.00 per 1M tokens blended. Choose Gemini 4 Carbon for reasoning (79 vs 59), knowledge (76 vs 62), and coding (72 vs 58), where it also streams ~2× faster; choose Kimi K3.1 Pro for fast first answers (55.9 s vs 103.1 s) and long context (68 vs 65).
Key facts (as of 29 Oct 2026)
- Gemini 4 Carbon scores 65.1 on the BenchLeader Index and Kimi K3.1 Pro scores 61.3 (BenchLeader, 29 Oct 2026).
- Gemini 4 Carbon costs $8.00 per 1M tokens blended (list price, 29 Oct 2026).
- Kimi K3.1 Pro costs $4.00 per 1M tokens blended (list price, 29 Oct 2026).
- Gemini 4 Carbon streams ~2× faster than Kimi K3.1 Pro (Artificial Analysis, 7-day median, 29 Oct 2026).
- Of the 17 benchmarks both models have, Gemini 4 Carbon leads 13 (BenchLeader, 29 Oct 2026).
Gemini 4 Carbon benchmarks high Google
Kimi K3.1 Pro benchmarks Moonshot AIOpen in the comparison toolWhich to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Gemini 4 Carbon | 72 vs 58 category score; FrontierCode 2.0 (AA) 40.6% vs 25.4% | High |
| Reasoning | Gemini 4 Carbon | 79 vs 59 category score; HLE-Diamond (AA) 69.1% vs 37.3% | High |
| Knowledge | Gemini 4 Carbon | 76 vs 62 category score; AA-Omniscience v1.1 53.5 vs 29.9 | High |
| Long documents | Kimi K3.1 Pro | 68 vs 65 category score; AA-LCR v1.1 86.3% vs 79.3% | Medium |
| Lowest cost | Kimi K3.1 Pro | $4.00 vs $8.00 per 1M blended; agentic coding session $0.160 vs $0.320 | – |
| Fastest first answer | Kimi K3.1 Pro | 55.9 s (Kimi K3.1 Pro) vs 103.1 s (Gemini 4 Carbon (high)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Gemini 4 Carbon (high) and Kimi K3.1 Pro are neck and neck on quality (65.1 vs 61.3, within the ±7.2 margin).
- Gemini 4 Carbon (high) is stronger in coding, knowledge, reasoning.
- Kimi K3.1 Pro is stronger in long context, agents & tools, composite, maths, multimodal.
- Kimi K3.1 Pro is 2.0× cheaper ($4.00 vs $8.00 per 1M blended).
- Gemini 4 Carbon (high) streams 2.6× faster (100 vs 38 tokens per second).
At a glance
| Spec | Gemini 4 Carbon | Kimi K3.1 Pro |
|---|---|---|
| Developer | Moonshot AI | |
| Released | 28 Oct 2026 | 23 Oct 2026 |
| Context window (tokens) | 1,000,000 | 1,048,576 |
| Compared at | Gemini 4 Carbon (high) | Kimi K3.1 Pro |
| Open weights | No | No |
| Price in / out / cached input per 1M | $4.00 / $20.00 / – | $2.00 / $10.00 / $0.200 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Gemini 4 Carbon (high) | Kimi K3.1 Pro | Leader |
|---|---|---|---|
| Reasoning | 78.9 | 59.0 | Gemini 4 Carbon |
| Coding | 71.7 | 58.1 | Gemini 4 Carbon |
| Agents & tools | – | 68.0 | not comparable |
| Maths | – | 73.7 | not comparable |
| Knowledge | 75.8 | 62.0 | Gemini 4 Carbon |
| Multimodal | – | 66.7 | not comparable |
| Long context | 64.5 | 68.1 | Kimi K3.1 Pro |
| Composite | – | 56.9 | not comparable |
Where sources disagree
Across the 17 benchmarks both have, Gemini 4 Carbon leads 13, Kimi K3.1 Pro leads 4. Every overall index that rates both puts Gemini 4 Carbon ahead.
| Overall index | Gemini 4 Carbon (high) | Kimi K3.1 Pro | Leader |
|---|---|---|---|
| BenchLeader Index | 65.1 | 61.3 | neck and neck |
| AA Intelligence Index v5.0 | 59.4 | 46 | Gemini 4 Carbon |
| Vals Index | 70.4 | 62.5† | Gemini 4 Carbon |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | Gemini 4 Carbon (high) | Kimi K3.1 Pro |
|---|---|---|
| Blended price per 1M tokens | $8.00 | $4.00 |
| Input / output per 1M | $4.00 / $20.00 | $2.00 / $10.00 |
| Cached input per 1M | – | $0.200 |
| Price per task: chat reply | $0.0076 | $0.0038 |
| Price per task: summarize a 30-page report | $0.060 | $0.030 |
| Price per task: code edit | $0.054 | $0.027 |
| Price per task: agentic coding session | $0.320 | $0.160 |
| Price per task: structured extraction | $0.012 | $0.0060 |
| Cost to run a full benchmark suite | $7.76 | $3.68 |
| Output speed | 100 tok/s | 38 tok/s |
| Time to first answer | 103.1 s | 55.9 s |
| Time to a complete answer | 108.1 s | 69.1 s |
Per token is not per task: on Artificial Analysis' full suite Gemini 4 Carbon (high) cost $7.76 and Kimi K3.1 Pro $3.68, because Gemini 4 Carbon (high) used about 1.1× as many tokens for the same work.
Each model is compared at its best-ranked reasoning-effort setting. Gemini 4 Carbon (high) scores 65.1 and takes 103 s to its first answer token. Kimi K3.1 Pro scores 61.3 and takes 56 s to its first answer token. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
56 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Gemini 4 Carbon (high) | Kimi K3.1 Pro |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.1 | 61.3 |
| Coding score | BenchLeader | 71.7 | 58.1 |
| Knowledge score | BenchLeader | 75.8 | 62.0 |
| Long context score | BenchLeader | 64.5 | 68.1 |
| Reasoning score | BenchLeader | 78.9 | 59.0 |
| Agents & tools score | BenchLeader | – | 68.0 |
| Composite score | BenchLeader | – | 56.9 |
| Maths score | BenchLeader | – | 73.7 |
| Multimodal score | BenchLeader | – | 66.7 |
| Blended price $/M | BenchLeader | $8.00 | $4.00 |
| Output speed | BenchLeader | 100 tok/s | 38 tok/s |
| Time to first answer | BenchLeader | 103.1 s | 55.9 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| LiveBench 2.0 Reasoning | LiveBench | – | 91.1% |
| Humanity's Last Exam (AA) | Artificial Analysis | 60.4% | 52.5% |
| HLE-Diamond (AA) | Artificial Analysis | 69.1% | 37.3% |
| CritPt | Artificial Analysis | 36.0% | 24.6% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 63.0% | 45.0% |
| MysteryMechanism | Vals AI | – | 26.1%† |
| Coding | |||
| SciCode | SciCode | 64.2% | 60.8% |
| Arena WebDev | Arena (LMArena) | – | 1695† |
| LiveBench 2.0 Coding | LiveBench | – | 81.8% |
| IOI | Vals AI | – | 81.9%† |
| Vibe Code Bench v1.1 | Vals AI | 89.1% | 84.9%† |
| ALE-Bench | Epoch AI Benchmarking Hub | – | 1932.8† |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 65.2% | 67.2% |
| Startrise Frontend | Startrise | – | 82.5% |
| FrontierCode 1.1 (AA) | Artificial Analysis | 48.5% | 47.1% |
| FrontierCode 2.0 (AA) | Artificial Analysis | 40.6% | 25.4% |
| Agents & tools | |||
| LiveBench Long Agentic | LiveBench | – | 39.1% |
| LiveBench Browser | LiveBench | – | 75.9% |
| GDPval-AA v2.1 | Artificial Analysis | – | 67.3% |
| Analyst Agent (AA) | Artificial Analysis | – | 38.8% |
| Finance Agent v2.1 | Vals AI | 67.2% | 61.5%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 17.5% | 11.7%† |
| Legal Research Bench | Vals AI | 55.8% | 37.5%† |
| Tax Agent Bench | Vals AI | 48.9% | 24.9%† |
| Terminal-Bench 4.0 (Vals) | Vals AI | 63.1% | 63.6%† |
Show all 56 benchmarks
| Metric | Publisher | Gemini 4 Carbon (high) | Kimi K3.1 Pro |
|---|---|---|---|
| Agents & tools | |||
| Terminal-Bench Science | Vals AI | – | 31.4%† |
| Vending-Bench 2 | Andon Labs | 16484.2† | – |
| AutomationBench | Artificial Analysis | 77.8% | 73.3% |
| GDP.pdf (AA) | Artificial Analysis | 28.4% | 31.4% |
| AA-Briefcase v1.1 | Artificial Analysis | 1609 | 1678 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 64.8% | 34.3% |
| CWE-Bench (AA) | Artificial Analysis | – | 60.8% |
| DeepSec-Bench (AA) | Artificial Analysis | – | 36.7% |
| CyberGym end-to-end (AA) | Artificial Analysis | – | 94.7% |
| Maths | |||
| ProofBench | Vals AI | – | 91.0% |
| Knowledge | |||
| LiveBench 2.0 Data Analysis | LiveBench | – | 79.1% |
| LiveBench Research | LiveBench | – | 77.9% |
| AA-Omniscience v1.1 | Artificial Analysis | 53.5 | 29.9 |
| BioMysteryBench | Vals AI | – | 75.6%† |
| Excel Modeling Benchmark v1.1 | Vals AI | 76.5% | 70.0%† |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 58.3% | 48.9% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 88.6% | 62.7% |
| Multimodal | |||
| LiveBench Design | LiveBench | – | 77.8% |
| MMMU-Pro | Artificial Analysis | – | 83.9% |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 58.6%† | – |
| AA-3D | Artificial Analysis | – | 1110 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 79.3% | 86.3% |
| MLCR | Artificial Analysis | – | 38.3% |
| Composite | |||
| LiveBench 2.0 | LiveBench | – | 71.3% |
| AA Intelligence Index v5.0 | Artificial Analysis | 59.4 | 46 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | – | 55.1 |
| AA Progress Index v1 | Artificial Analysis | 351.6 | 340.7 |
| Vals Index | Vals AI | 70.4 | 62.5† |
| Vals Finance Index | Vals AI | 66.3% | 56.0%† |
| Vals Science Index | Vals AI | – | 44.4%† |
| AA Cyber Index | Artificial Analysis | – | 64.1 |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- Kimi K3.1 Pro: BenchLeader Index 59.3 → 61.3.
- Kimi K3.1 Pro: 3 new results (SciCode, ProofBench, Startrise Frontend).
- Kimi K3.1 Pro: BenchLeader Index 61.8 → 59.3.
- Kimi K3.1 Pro: 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 2 score revisions; Kimi K3.1 Pro cost per task moved from $2.11/M to $3.68/M (74% up).
- Kimi K3.1 Pro: BenchLeader Index 62.3 → 61.8.
- Kimi K3.1 Pro: 1 new result (MLCR); 3 score revisions.
- Kimi K3.1 Pro: 3 new results (Analyst Agent (AA), Terminal-Bench-Science 0.1 (AA), AA-3D).
- Kimi K3.1 Pro: 5 new results (FrontierCode 1.1 (AA), AA Cyber Index, CWE-Bench (AA)…).
- Kimi K3.1 Pro: first results.
Related comparisons
- Gemini 4 Carbon vs Claude Opus 5.5
- Gemini 4 Carbon vs GPT-6 Astra
- Gemini 4 Carbon vs GPT-6.1 Sol
- Kimi K3.1 Pro vs Claude Opus 5.5
- Kimi K3.1 Pro vs GPT-6 Astra
- Kimi K3.1 Pro vs GPT-6.1 Sol
- Kimi K3.1 Pro vs Claude Fable 5.1
- Gemini 4 Argon vs Kimi K3.1 Pro
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Gemini 4 Carbon vs Kimi K3.1 Pro: questions
- Is Gemini 4 Carbon better than Kimi K3.1 Pro?
- Gemini 4 Carbon (high) and Kimi K3.1 Pro are neck and neck on quality (65.1 vs 61.3, within the ±7.2 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 29 Oct 2026, so check the category scores for your use.
- Gemini 4 Carbon vs Kimi K3.1 Pro: which is cheaper, and will that change?
- Kimi K3.1 Pro is cheaper today: $4.00 against $8.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Gemini 4 Carbon or Kimi K3.1 Pro better for coding?
- Gemini 4 Carbon scores higher in coding (72 vs 58 on the category index, where 50 is average). On FrontierCode 2.0 (AA), the top shared coding benchmark, it is 40.6% for Gemini 4 Carbon against 25.4% for Kimi K3.1 Pro.
- Which is faster, Gemini 4 Carbon or Kimi K3.1 Pro?
- Gemini 4 Carbon streams ~2× faster (100 vs 38 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- Kimi K3.1 Pro scores higher on long-context benchmarks (68 vs 65). Both take about 1M tokens: Gemini 4 Carbon 1,000,000, Kimi K3.1 Pro 1,048,576.
More questions (3)
- Which has lower latency?
- Kimi K3.1 Pro answers first: 55.9 s against 103.1 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Gemini 4 Carbon vs Kimi K3.1 Pro?
- Across the 17 benchmarks both have, Gemini 4 Carbon leads 13, Kimi K3.1 Pro leads 4. Every overall index that rates both puts Gemini 4 Carbon ahead.
- When was Gemini 4 Carbon released?
- Gemini 4 Carbon was released on 28 Oct 2026.