Gemini 4 Carbon vs Grok 4.6
Full profiles: Gemini 4 Carbon (Google) and Grok 4.6 (SpaceXAI): every benchmark, price and rank.
Updated · Data from 8 sources
As of 29 Oct 2026, Gemini 4 Carbon and Grok 4.6 are neck and neck on quality (65.1 vs 60.6 on the BenchLeader Index, inside the ±7.0 margin), and Grok 4.6 is 2.7× cheaper at $3.00 vs $8.00 per 1M tokens blended. Gemini 4 Carbon leads in every category we measure both in.
Key facts (as of 29 Oct 2026)
- Gemini 4 Carbon scores 65.1 on the BenchLeader Index and Grok 4.6 scores 60.6 (BenchLeader, 29 Oct 2026).
- Gemini 4 Carbon costs $8.00 per 1M tokens blended (list price, 29 Oct 2026).
- Grok 4.6 costs $3.00 per 1M tokens blended (list price, 29 Oct 2026).
- Of the 14 benchmarks both models have, Gemini 4 Carbon leads 13 (BenchLeader, 29 Oct 2026).
Gemini 4 Carbon benchmarks 
Grok 4.6 benchmarks Which 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 61 category score; SciCode 64.2% vs 56.5% | High |
| Reasoning | Gemini 4 Carbon | 79 vs 64 category score; Humanity's Last Exam (AA) 60.4% vs 42.9% | High |
| Knowledge | Gemini 4 Carbon | 76 vs 60 category score | High |
| Long documents | Either | 65 vs 65 category score; AA-LCR v1.1 79.3% vs 80.3% | Either |
| Chat quality (human preference) | – | Only Grok 4.6 has an Arena rating (64 category score) | Not comparable |
| Lowest cost | Grok 4.6 | $3.00 vs $8.00 per 1M blended; agentic coding session $0.144 vs $0.320 | – |
| Fastest first answer | Grok 4.6 | 27.1 s (Grok 4.6 (high)) 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 Grok 4.6 (high) are neck and neck on quality (65.1 vs 60.6, within the ±7.0 margin).
- Gemini 4 Carbon (high) is stronger in coding, knowledge, reasoning.
- Grok 4.6 (high) is stronger in long context, agents & tools, human preference, maths, multimodal.
- Grok 4.6 (high) is 2.7× cheaper ($3.00 vs $8.00 per 1M blended).
At a glance
| Spec | Gemini 4 Carbon | Grok 4.6 |
|---|---|---|
| Developer | SpaceXAI | |
| Released | 28 Oct 2026 | 11 Aug 2026 |
| Context window (tokens) | 1,000,000 | 500,000 |
| Compared at | Gemini 4 Carbon (high) | Grok 4.6 (high) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $4.00 / $20.00 / – | $2.00 / $6.00 / $0.500 |
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) | Grok 4.6 (high) | Leader |
|---|---|---|---|
| Reasoning | 78.9 | 64.0 | Gemini 4 Carbon |
| Coding | 71.7 | 61.4 | Gemini 4 Carbon |
| Agents & tools | – | 58.1 | not comparable |
| Maths | – | 63.5 | not comparable |
| Knowledge | 75.8 | 59.6 | Gemini 4 Carbon |
| Human preference | – | 64.2 | not comparable |
| Multimodal | – | 59.8 | not comparable |
| Long context | 64.5 | 65.0 | level |
Where sources disagree
Across the 14 benchmarks both have, Gemini 4 Carbon leads 13, Grok 4.6 leads 1.
| Overall index | Gemini 4 Carbon (high) | Grok 4.6 (high) | Leader |
|---|---|---|---|
| BenchLeader Index | 65.1 | 60.6 | neck and neck |
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) | Grok 4.6 (high) |
|---|---|---|
| Blended price per 1M tokens | $8.00 | $3.00 |
| Input / output per 1M | $4.00 / $20.00 | $2.00 / $6.00 |
| Cached input per 1M | – | $0.500 |
| Price per task: chat reply | $0.0076 | $0.0026 |
| Price per task: summarize a 30-page report | $0.060 | $0.028 |
| Price per task: code edit | $0.054 | $0.021 |
| Price per task: agentic coding session | $0.320 | $0.144 |
| Price per task: structured extraction | $0.012 | $0.0052 |
| Cost to run a full benchmark suite | $7.76 | – |
| Output speed | 100 tok/s | 88 tok/s |
| Time to first answer | 103.1 s | 27.1 s |
| Time to a complete answer | 108.1 s | 32.7 s |
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. Grok 4.6 (high) scores 60.6 and takes 27 s to its first answer token; Grok 4.6 (medium) scores 59.6 in 23 s, Grok 4.6 (xhigh) scores 58.5 in 34 s. 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
119 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) | Grok 4.6 (high) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.1 | 60.6 |
| Coding score | BenchLeader | 71.7 | 61.4 |
| Knowledge score | BenchLeader | 75.8 | 59.6 |
| Long context score | BenchLeader | 64.5 | 65.0 |
| Reasoning score | BenchLeader | 78.9 | 64.0 |
| Agents & tools score | BenchLeader | – | 58.1 |
| Human preference score | BenchLeader | – | 64.2 |
| Maths score | BenchLeader | – | 63.5 |
| Multimodal score | BenchLeader | – | 59.8 |
| Blended price $/M | BenchLeader | $8.00 | $3.00 |
| Output speed | BenchLeader | 100 tok/s | 88 tok/s |
| Time to first answer | BenchLeader | 103.1 s | 27.1 s |
| Context window | BenchLeader | 1,000,000 tokens | 500,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | – | 94.0% |
| SimpleBench | SimpleBench | – | 75.9%† |
| Arena Hard Prompts | Arena (LMArena) | – | 1482 |
| LiveBench 2.0 Reasoning | LiveBench | – | 91.2%† |
| LiveBench 1.0 Reasoning | LiveBench | – | 90.5%† |
| GPQA Diamond (AA) | Artificial Analysis | – | 95.0% |
| Humanity's Last Exam (AA) | Artificial Analysis | 60.4% | 42.9% |
| HLE-Diamond (AA) | Artificial Analysis | 69.1% | – |
| GPQA Diamond (Vals) | Vals AI | – | 94.7% |
| ARC-AGI-1 | ARC Prize | – | 87.0% |
| ARC-AGI-2 | ARC Prize | – | 65.1% |
| ARC-AGI-3 | ARC Prize | – | 2.1%† |
| CritPt | Artificial Analysis | 36.0% | 17.1% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 63.0% | – |
| MysteryMechanism | Vals AI | – | 30.6% |
| Chess Puzzles | Epoch AI Benchmarking Hub | – | 33.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | – | 34.0%† |
| LMCA | Epoch AI Benchmarking Hub | – | 48.5%† |
| DTBench | Epoch AI Benchmarking Hub | – | 97.3%† |
| ForecastBench | ForecastBench | – | 65.5%† |
| Coding | |||
| SciCode | SciCode | 64.2% | 56.5% |
| WeirdML | WeirdML | – | 67.3% |
| FrontierCode | Cognition | – | 48.0%† |
| Arena Coding | Arena (LMArena) | – | 1510 |
| Arena WebDev | Arena (LMArena) | – | 1617 |
Show all 119 benchmarks
| Metric | Publisher | Gemini 4 Carbon (high) | Grok 4.6 (high) |
|---|---|---|---|
| Coding | |||
| LiveBench 2.0 Coding | LiveBench | – | 89.1%† |
| LiveBench 1.0 Coding | LiveBench | – | 76.8%† |
| LiveCodeBench | Vals AI | – | 88.2% |
| IOI | Vals AI | – | 47.6% |
| SWE-bench (Vals) | Vals AI | – | 95.6% |
| Code Migration | Vals AI | – | 44.6% |
| ProgramBench | Vals AI | – | 1.0% |
| Vibe Code Bench 1-100 | Vals AI | – | 14.8% |
| Vibe Code Bench v1.1 | Vals AI | 89.1% | 76.2% |
| MirrorCode | Epoch AI Benchmarking Hub | – | 41.1%† |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | – | 65.2% |
| CursorBench | Cursor | – | 40.4% |
| ALE-Bench | Epoch AI Benchmarking Hub | – | 1508.1† |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 25.3%† |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 65.2% | 21.2% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 88.4% |
| Startrise Frontend | Startrise | – | 72.1%† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 48.5% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 40.6% | – |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | – | 20.3% |
| APEX-Agents | Mercor | – | 65.3%† |
| Arena Agent | Arena (LMArena) | – | 0.6† |
| LiveBench Long Agentic | LiveBench | – | 28.0%† |
| LiveBench Browser | LiveBench | – | 49.4%† |
| LiveBench Agentic Coding | LiveBench | – | 57.0%† |
| Terminal-Bench 2.1 (Vals) | Vals AI | – | 78.3% |
| τ³-Banking (AA) | Artificial Analysis | – | 50.7% |
| Analyst Agent (AA) | Artificial Analysis | – | 41.3% |
| CyberBench | Vals AI | – | 67.7% |
| Finance Agent v2 | Vals AI | – | 53.7% |
| Finance Agent v2.1 | Vals AI | 67.2% | – |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 17.5% | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | – | 15.8% |
| Legal Research Bench | Vals AI | 55.8% | 48.1% |
| SkillsBench | Vals AI | – | 55.8% |
| Tax Agent Bench | Vals AI | 48.9% | 33.1% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 63.1% | 17.2% |
| Terminal-Bench Science | Vals AI | – | 11.4% |
| Time Horizon Index: KSP | Vals AI | – | 5.8% |
| Vending-Bench 2 | Andon Labs | 16484.2† | 9047.0† |
| AutomationBench | Artificial Analysis | 77.8% | 66.7% |
| GDP.pdf (AA) | Artificial Analysis | 28.4% | 17.0% |
| EnterpriseOps-Gym | Artificial Analysis | – | 48.3% |
| AA-Briefcase v1.1 | Artificial Analysis | 1609 | 1529 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 64.8% | 6.2% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | – | 66.0%† |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | – | 31.7%† |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | – | 97.8% |
| ProofBench | Vals AI | – | 51.0%† |
| LiveBench Mathematics | LiveBench | – | 92.6%† |
| Arena Maths | Arena (LMArena) | – | 1445 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | – | 49.3% |
| LiveBench 2.0 Data Analysis | LiveBench | – | 75.3%† |
| LiveBench Research | LiveBench | – | 84.5%† |
| LiveBench 1.0 Data Analysis | LiveBench | – | 73.9%† |
| AA-Omniscience v1.1 | Artificial Analysis | 53.5 | – |
| MMLU-Pro | Vals AI | – | 89.4% |
| LegalBench | Vals AI | – | 86.3% |
| CorpFin | Vals AI | – | 66.2% |
| TaxEval | Vals AI | – | 71.1% |
| BioMysteryBench | Vals AI | – | 72.2% |
| Excel Modeling Benchmark | Vals AI | – | 62.7% |
| Excel Modeling Benchmark v1.1 | Vals AI | 76.5% | – |
| MedCode | Vals AI | – | 44.7% |
| MedScribe | Vals AI | – | 86.5% |
| Public Benefits Bench v1.1 | Vals AI | – | 66.8% |
| EBR-bench | Epoch AI Benchmarking Hub | – | 30.5%† |
| GDP.pdf | Epoch AI Benchmarking Hub | – | 16.0% |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 58.3% | – |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 88.6% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 83.7%† |
| LiveBench Instruction Following | LiveBench | – | 71.9%† |
| Arena Instruction Following | Arena (LMArena) | – | 1452 |
| Human preference | |||
| Arena Text | Arena (LMArena) | – | 1456 |
| Arena Creative Writing | Arena (LMArena) | – | 1443 |
| Arena Multi-turn | Arena (LMArena) | – | 1448 |
| Arena Longer Queries | Arena (LMArena) | – | 1475 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | – | 1263 |
| LiveBench Design | LiveBench | – | 79.8%† |
| MortgageTax | Vals AI | – | 64.2% |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 58.6%† | 33.2%† |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 79.3% | 80.3% |
| Arena Document | Arena (LMArena) | – | 1461 |
| MLCR | Artificial Analysis | – | 15.6% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | – | 156.6† |
| LiveBench 2.0 | LiveBench | – | 67.8%† |
| LiveBench 1.0 | LiveBench | – | 78.0%† |
| AA Intelligence Index v5.0 | Artificial Analysis | 59.4 | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | – | 44.3 |
| AA Progress Index v1 | Artificial Analysis | 351.6 | 321.5 |
| Vals Index | Vals AI | 70.4 | – |
| Vals Finance Index | Vals AI | 66.3% | – |
| Vals RSI Index | Vals AI | – | 25.5%† |
| Vals Science Index | Vals AI | – | 38.1% |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- Grok 4.6 (high): BenchLeader Index 61.1 → 60.6.
- Grok 4.6 (high): 1 new result (AA Progress Index v1).
- Grok 4.6 (high): 2 new results (Harvey's Legal Agent Benchmark v1, Vals Science Index).
- Grok 4.6 (high): 1 score revision.
- Grok 4.6 (high): BenchLeader Index 61.6 → 61.1.
- Grok 4.6 (high): 3 new results (GDPval-AA v2.1, AA-Omniscience v1.1: accuracy, AA-Omniscience v1.1: non-hallucination).
- Grok 4.6 (high): 1 new result (ProgramBench); 1 score revision.
- Grok 4.6 (high): 1 score revision.
- Grok 4.6 (high): 1 new result (Terminal-Bench-Science 0.1 (AA)).
- Grok 4.6 (high): 1 score revision; cost per task moved from 1.9 to 1.5 (20% down).
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
- Grok 4.6 vs Claude Opus 5.5
- Grok 4.6 vs GPT-6 Astra
- Grok 4.6 vs GPT-6.1 Sol
- Grok 4.6 vs Claude Fable 5.1
- Gemini 4 Argon vs Grok 4.6
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Gemini 4 Carbon vs Grok 4.6: questions
- Is Gemini 4 Carbon better than Grok 4.6?
- Gemini 4 Carbon (high) and Grok 4.6 (high) are neck and neck on quality (65.1 vs 60.6, within the ±7.0 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 Grok 4.6: which is cheaper, and will that change?
- Grok 4.6 is cheaper today: $3.00 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 Grok 4.6 better for coding?
- Gemini 4 Carbon scores higher in coding (72 vs 61 on the category index, where 50 is average). On SciCode, the top shared coding benchmark, it is 64.2% for Gemini 4 Carbon against 56.5% for Grok 4.6.
- Which is faster, Gemini 4 Carbon or Grok 4.6?
- Gemini 4 Carbon streams faster (100 vs 88 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- They are level on long-context benchmarks (65 vs 65). Gemini 4 Carbon accepts more context: 1,000,000 against 500,000 tokens for Grok 4.6.
More questions (3)
- Which has lower latency?
- Grok 4.6 answers first: 27.1 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 Grok 4.6?
- Across the 14 benchmarks both have, Gemini 4 Carbon leads 13, Grok 4.6 leads 1.
- When was Gemini 4 Carbon released?
- Gemini 4 Carbon was released on 28 Oct 2026.