Gemini 4 Carbon vs Muse Spark 1.3
Full profiles: Gemini 4 Carbon (Google) and Muse Spark 1.3 (Meta): every benchmark, price and rank.
Updated · Data from 6 sources
As of 29 Oct 2026, Gemini 4 Carbon and Muse Spark 1.3 are neck and neck on quality (65.1 vs 61.7 on the BenchLeader Index, inside the ±7.1 margin), and Muse Spark 1.3 is 4.0× cheaper at $2.00 vs $8.00 per 1M tokens blended. Choose Gemini 4 Carbon for reasoning (79 vs 60), knowledge (76 vs 58), and coding (72 vs 60); choose Muse Spark 1.3 for fast first answers (46.6 s vs 103.1 s) and long context (66 vs 65), where it also streams ~2× faster.
Key facts (as of 29 Oct 2026)
- Gemini 4 Carbon scores 65.1 on the BenchLeader Index and Muse Spark 1.3 scores 61.7 (BenchLeader, 29 Oct 2026).
- Gemini 4 Carbon costs $8.00 per 1M tokens blended (list price, 29 Oct 2026).
- Muse Spark 1.3 costs $2.00 per 1M tokens blended (list price, 29 Oct 2026).
- Muse Spark 1.3 streams ~2× faster than Gemini 4 Carbon (Artificial Analysis, 7-day median, 29 Oct 2026).
- Of the 21 benchmarks both models have, Gemini 4 Carbon leads 20 (BenchLeader, 29 Oct 2026).
Gemini 4 Carbon benchmarks high Google
Muse Spark 1.3 benchmarks max MetaOpen 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 60 category score; FrontierCode 2.0 (AA) 40.6% vs 23.1% | High |
| Reasoning | Gemini 4 Carbon | 79 vs 60 category score; HLE-Diamond (AA) 69.1% vs 33.7% | High |
| Knowledge | Gemini 4 Carbon | 76 vs 58 category score; AA-Omniscience v1.1 53.5 vs 22.4 | High |
| Long documents | Muse Spark 1.3 | 66 vs 65 category score; AA-LCR v1.1 83.0% vs 79.3% | Medium |
| Chat quality (human preference) | – | Only Muse Spark 1.3 has an Arena rating (69 category score) | Not comparable |
| Lowest cost | Muse Spark 1.3 | $2.00 vs $8.00 per 1M blended; agentic coding session $0.092 vs $0.320 | – |
| Fastest first answer | Muse Spark 1.3 | 46.6 s (Muse Spark 1.3 (max)) 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 Muse Spark 1.3 (max) are neck and neck on quality (65.1 vs 61.7, within the ±7.1 margin).
- Gemini 4 Carbon (high) is stronger in coding, knowledge, reasoning.
- Muse Spark 1.3 (max) is stronger in long context, agents & tools, human preference, maths, multimodal.
- Muse Spark 1.3 (max) is 4.0× cheaper ($2.00 vs $8.00 per 1M blended).
- Muse Spark 1.3 (max) streams 2.1× faster (208 vs 100 tokens per second).
At a glance
| Spec | Gemini 4 Carbon | Muse Spark 1.3 |
|---|---|---|
| Developer | Meta | |
| Released | 28 Oct 2026 | 2 Sept 2026 |
| Context window (tokens) | 1,000,000 | 1,048,576 |
| Compared at | Gemini 4 Carbon (high) | Muse Spark 1.3 (max) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $4.00 / $20.00 / – | $1.25 / $4.25 / $0.150 |
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) | Muse Spark 1.3 (max) | Leader |
|---|---|---|---|
| Reasoning | 78.9 | 60.0 | Gemini 4 Carbon |
| Coding | 71.7 | 60.2 | Gemini 4 Carbon |
| Agents & tools | – | 65.7 | not comparable |
| Maths | – | 63.2 | not comparable |
| Knowledge | 75.8 | 57.7 | Gemini 4 Carbon |
| Human preference | – | 69.0 | not comparable |
| Multimodal | – | 65.7 | not comparable |
| Long context | 64.5 | 66.4 | Muse Spark 1.3 |
Where sources disagree
Across the 21 benchmarks both have, Gemini 4 Carbon leads 20, Muse Spark 1.3 leads 1. Every overall index that rates both puts Gemini 4 Carbon ahead.
| Overall index | Gemini 4 Carbon (high) | Muse Spark 1.3 (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 65.1 | 61.7 | neck and neck |
| AA Intelligence Index v5.0 | 59.4 | 36.4 | 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) | Muse Spark 1.3 (max) |
|---|---|---|
| Blended price per 1M tokens | $8.00 | $2.00 |
| Input / output per 1M | $4.00 / $20.00 | $1.25 / $4.25 |
| Cached input per 1M | – | $0.150 |
| Price per task: chat reply | $0.0076 | $0.0018 |
| Price per task: summarize a 30-page report | $0.060 | $0.018 |
| Price per task: code edit | $0.054 | $0.014 |
| Price per task: agentic coding session | $0.320 | $0.092 |
| Price per task: structured extraction | $0.012 | $0.0034 |
| Cost to run a full benchmark suite | $7.76 | $2.39 |
| Output speed | 100 tok/s | 208 tok/s |
| Time to first answer | 103.1 s | 46.6 s |
| Time to a complete answer | 108.1 s | 49.0 s |
Per token is not per task: on Artificial Analysis' full suite Gemini 4 Carbon (high) cost $7.76 and Muse Spark 1.3 (max) $2.39, because Muse Spark 1.3 (max) used about 1.2× 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. Muse Spark 1.3 (max) scores 61.7 and takes 47 s to its first answer token; Muse Spark 1.3 (xhigh) scores 58.4 in 25 s, Muse Spark 1.3 scores 57.9 in 2 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
103 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) | Muse Spark 1.3 (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.1 | 61.7 |
| Coding score | BenchLeader | 71.7 | 60.2 |
| Knowledge score | BenchLeader | 75.8 | 57.7 |
| Long context score | BenchLeader | 64.5 | 66.4 |
| Reasoning score | BenchLeader | 78.9 | 60.0 |
| Agents & tools score | BenchLeader | – | 65.7 |
| Human preference score | BenchLeader | – | 69.0 |
| Maths score | BenchLeader | – | 63.2 |
| Multimodal score | BenchLeader | – | 65.7 |
| Blended price $/M | BenchLeader | $8.00 | $2.00 |
| Output speed | BenchLeader | 100 tok/s | 208 tok/s |
| Time to first answer | BenchLeader | 103.1 s | 46.6 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| Arena Hard Prompts | Arena (LMArena) | – | 1517 |
| LiveBench 2.0 Reasoning | LiveBench | – | 89.7%† |
| LiveBench 1.0 Reasoning | LiveBench | – | 89.7%† |
| GPQA Diamond (AA) | Artificial Analysis | – | 93.5% |
| Humanity's Last Exam (AA) | Artificial Analysis | 60.4% | 48.7% |
| HLE-Diamond (AA) | Artificial Analysis | 69.1% | 33.7% |
| CritPt | Artificial Analysis | 36.0% | 24.9% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 63.0% | 46.0% |
| MysteryMechanism | Vals AI | – | 36.0% |
| Chess Puzzles | Epoch AI Benchmarking Hub | – | 37.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | – | 25.0% |
| LMCA | Epoch AI Benchmarking Hub | – | 53.9%† |
| DTBench | Epoch AI Benchmarking Hub | – | 96.5%† |
| BTF-3 | Epoch AI Benchmarking Hub | – | 14.1%† |
| Coding | |||
| SciCode | SciCode | 64.2% | 58.8% |
| Arena Coding | Arena (LMArena) | – | 1536 |
| Arena WebDev | Arena (LMArena) | – | 1656 |
| LiveBench 2.0 Coding | LiveBench | – | 90.2%† |
| LiveBench 1.0 Coding | LiveBench | – | 81.1%† |
| IOI | Vals AI | – | 56.6% |
| Code Migration | Vals AI | – | 47.4% |
| ProgramBench | Vals AI | – | 2.5% |
| SWE-bench Multimodal | Vals AI | – | 43.1% |
| Vibe Code Bench 1-100 | Vals AI | – | 20.5% |
| Vibe Code Bench v1.1 | Vals AI | 89.1% | 85.9% |
Show all 103 benchmarks
| Metric | Publisher | Gemini 4 Carbon (high) | Muse Spark 1.3 (max) |
|---|---|---|---|
| Coding | |||
| CursorBench | Cursor | – | 41.6% |
| ALE-Bench | Epoch AI Benchmarking Hub | – | 1688.3† |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 25.8% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 65.2% | 33.3% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 84.3% |
| Startrise Frontend | Startrise | – | 74.7%† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 48.5% | 40.6% |
| FrontierCode 2.0 (AA) | Artificial Analysis | 40.6% | 23.1% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | – | 14.6%† |
| APEX-Agents | Mercor | – | 47.6% |
| Arena Agent | Arena (LMArena) | – | 3.1 |
| LiveBench Long Agentic | LiveBench | – | 25.5%† |
| LiveBench Browser | LiveBench | – | 30.9%† |
| LiveBench Agentic Coding | LiveBench | – | 64.1%† |
| Terminal-Bench 2.1 (Vals) | Vals AI | – | 79.0% |
| HiL-Bench | Scale AI SEAL | – | 66.9%† |
| τ³-Banking (AA) | Artificial Analysis | – | 50.5% |
| ITBench SRE (AA) | Artificial Analysis | – | 33.2% |
| Analyst Agent (AA) | Artificial Analysis | – | 37.5% |
| CUA-bench | Vals AI | – | 5.8% |
| CyberBench | Vals AI | – | 72.7% |
| Finance Agent v2 | Vals AI | – | 60.0% |
| 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 | – | 23.8% |
| Legal Research Bench | Vals AI | 55.8% | 55.3% |
| SREBench | Vals AI | – | 3.8% |
| Tax Agent Bench | Vals AI | 48.9% | 42.4% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 63.1% | 24.8% |
| Terminal-Bench Science | Vals AI | – | 10.0% |
| Vending-Bench 2 | Andon Labs | 16484.2† | – |
| AutomationBench | Artificial Analysis | 77.8% | 57.9% |
| GDP.pdf (AA) | Artificial Analysis | 28.4% | 26.6% |
| Harvey LAB-AA v1.1 | Harvey | – | 8.9% |
| AA-Briefcase v1.1 | Artificial Analysis | 1609 | 1581 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 64.8% | 10.9% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | – | 74.0% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | – | 46.3% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | – | 99.2%† |
| ProofBench | Vals AI | – | 58.0% |
| LiveBench Mathematics | LiveBench | – | 96.0%† |
| Arena Maths | Arena (LMArena) | – | 1510 |
| Knowledge | |||
| LiveBench 2.0 Data Analysis | LiveBench | – | 79.6%† |
| LiveBench Research | LiveBench | – | 68.6%† |
| LiveBench 1.0 Data Analysis | LiveBench | – | 79.6%† |
| AA-Omniscience v1.1 | Artificial Analysis | 53.5 | 22.4 |
| PRBench Finance | Scale AI SEAL | – | 59.5%† |
| PRBench Legal | Scale AI SEAL | – | 61.6%† |
| Excel Modeling Benchmark | Vals AI | – | 67.4% |
| Excel Modeling Benchmark v1.1 | Vals AI | 76.5% | – |
| DrugDiscoveryBench | Scale AI SEAL | – | 62.2%† |
| GDP.pdf | Epoch AI Benchmarking Hub | – | 27.6%† |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 58.3% | 38.0% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 88.6% | 74.8% |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 82.8%† |
| LiveBench Instruction Following | LiveBench | – | 78.0%† |
| Arena Instruction Following | Arena (LMArena) | – | 1485 |
| Human preference | |||
| Arena Text | Arena (LMArena) | – | 1496 |
| Arena Creative Writing | Arena (LMArena) | – | 1463 |
| Arena Multi-turn | Arena (LMArena) | – | 1494 |
| Arena Longer Queries | Arena (LMArena) | – | 1497 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | – | 1308 |
| LiveBench Design | LiveBench | – | 74.8%† |
| MMMU-Pro | Artificial Analysis | – | 82.0%† |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 58.6%† | – |
| AA-3D | Artificial Analysis | – | 1033 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 79.3% | 83.0% |
| Arena Document | Arena (LMArena) | – | 1468 |
| MLCR | Artificial Analysis | – | 13.9% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | – | 157.2† |
| LiveBench 2.0 | LiveBench | – | 63.0%† |
| LiveBench 1.0 | LiveBench | – | 81.6%† |
| AA Intelligence Index v5.0 | Artificial Analysis | 59.4 | 36.4 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | – | 48.1 |
| AA Progress Index v1 | Artificial Analysis | 351.6 | 324.5 |
| Vals Index | Vals AI | 70.4 | – |
| Vals Finance Index | Vals AI | 66.3% | – |
| Vals RSI Index | Vals AI | – | 19.6% |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- Muse Spark 1.3 (max): BenchLeader Index 64.6 → 61.8.
- Muse Spark 1.3 (max): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; cost per task moved from $1.31/M to $2.39/M (81% up).
- Muse Spark 1.3 (max): cost per task moved from $3.69/M to $1.31/M (64% down).
- Muse Spark 1.3 (max): cost per task moved from $1.60/M to $3.69/M (130% up).
- Muse Spark 1.3 (max): 3 new results (Harvey's Legal Agent Benchmark v1, Vals RSI Index, SWE-bench Multimodal); 1 score revision.
- Muse Spark 1.3 (max): 1 score revision.
- Muse Spark 1.3 (max): 1 score revision.
- Muse Spark 1.3 (max): 2 new results (ProgramBench, SREBench); 1 score revision.
- Muse Spark 1.3 (max): BenchLeader Index 65.2 → 64.6.
- Muse Spark 1.3 (max): 2 new results (APEX-Agents, FrontierSWE); 1 score revision.
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
- Muse Spark 1.3 vs Claude Opus 5.5
- Muse Spark 1.3 vs GPT-6 Astra
- Muse Spark 1.3 vs GPT-6.1 Sol
- Muse Spark 1.3 vs Claude Fable 5.1
- Gemini 4 Argon vs Muse Spark 1.3
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Gemini 4 Carbon vs Muse Spark 1.3: questions
- Is Gemini 4 Carbon better than Muse Spark 1.3?
- Gemini 4 Carbon (high) and Muse Spark 1.3 (max) are neck and neck on quality (65.1 vs 61.7, within the ±7.1 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 Muse Spark 1.3: which is cheaper, and will that change?
- Muse Spark 1.3 is cheaper today: $2.00 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 Muse Spark 1.3 better for coding?
- Gemini 4 Carbon scores higher in coding (72 vs 60 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 23.1% for Muse Spark 1.3.
- Which is faster, Gemini 4 Carbon or Muse Spark 1.3?
- Muse Spark 1.3 streams ~2× faster (208 vs 100 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- Muse Spark 1.3 scores higher on long-context benchmarks (66 vs 65). Both take about 1M tokens: Gemini 4 Carbon 1,000,000, Muse Spark 1.3 1,048,576.
More questions (3)
- Which has lower latency?
- Muse Spark 1.3 answers first: 46.6 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 Muse Spark 1.3?
- Across the 21 benchmarks both have, Gemini 4 Carbon leads 20, Muse Spark 1.3 leads 1. 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.