Claude Sonnet 5.5 vs Gemini 4 Hydrogen
Full profiles: Claude Sonnet 5.5 (Anthropic) and Gemini 4 Hydrogen (Google): every benchmark, price and rank.
Updated · Data from 10 sources
As of 13 Nov 2026, Claude Sonnet 5.5 and Gemini 4 Hydrogen are neck and neck on quality (64.7 vs 63.4 on the BenchLeader Index, inside the ±3.4 margin), and Gemini 4 Hydrogen is 1.3× cheaper at $3.00 vs $4.00 per 1M tokens blended. Choose Claude Sonnet 5.5 for agents & tools (71 vs 62), maths (71 vs 68), and coding (72 vs 69); choose Gemini 4 Hydrogen for fast first answers (16.4 s vs 247.0 s) and knowledge (69 vs 57) and reasoning (70 vs 68).
Key facts (as of 13 Nov 2026)
- Claude Sonnet 5.5 scores 64.7 on the BenchLeader Index and Gemini 4 Hydrogen scores 63.4 (BenchLeader, 13 Nov 2026).
- Claude Sonnet 5.5 costs $4.00 per 1M tokens blended (list price, 13 Nov 2026).
- Gemini 4 Hydrogen costs $3.00 per 1M tokens blended (list price, 13 Nov 2026).
Claude Sonnet 5.5 benchmarks max Anthropic
Gemini 4 Hydrogen benchmarks high GoogleAdd a third modelWhich 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 | Claude Sonnet 5.5 | 72 vs 69 category score; FrontierCode 2.0 (AA) 41.7% vs 36.3% | Medium |
| Agents & tool use | Claude Sonnet 5.5 | 71 vs 62 category score; Terminal-Bench 63.9% vs 56.4% | High |
| Maths | Claude Sonnet 5.5 | 71 vs 68 category score | Medium |
| Reasoning | Gemini 4 Hydrogen | 70 vs 68 category score; HLE-Diamond (AA) 54.9% vs 53.3% | Medium |
| Knowledge | Gemini 4 Hydrogen | 69 vs 57 category score; AA-Omniscience v1.1 43.7 vs 30.5 | High |
| Chat quality (human preference) | – | Only Gemini 4 Hydrogen has an Arena rating (68 category score) | Not comparable |
| Lowest cost | Gemini 4 Hydrogen | $3.00 vs $4.00 per 1M blended; agentic coding session $0.120 vs $0.160 | – |
| Fastest first answer | Gemini 4 Hydrogen | 16.4 s (Gemini 4 Hydrogen (high)) vs 247.0 s (Claude Sonnet 5.5 (max)) 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
- Claude Sonnet 5.5 (max) and Gemini 4 Hydrogen (high) are neck and neck on quality (64.7 vs 63.4, within the ±3.4 margin).
- Claude Sonnet 5.5 (max) is stronger in agents & tools, coding, maths.
- Gemini 4 Hydrogen (high) is stronger in knowledge, reasoning.
- Gemini 4 Hydrogen (high) is 1.3× cheaper ($3.00 vs $4.00 per 1M blended).
At a glance
| Spec | Claude Sonnet 5.5 | Gemini 4 Hydrogen |
|---|---|---|
| Developer | Anthropic | |
| Released | 28 Sep 2026 | 12 Nov 2026 |
| Context window (tokens) | 1,000,000 | 1,048,576 |
| Compared at | Claude Sonnet 5.5 (max) | Gemini 4 Hydrogen (high) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $2.00 / $10.00 / $0.100 | $1.50 / $7.50 / $0.075 |
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 | Claude Sonnet 5.5 (max) | Gemini 4 Hydrogen (high) | Leader |
|---|---|---|---|
| Reasoning | 67.7 | 69.6 | Gemini 4 Hydrogen |
| Coding | 71.5 | 68.8 | Claude Sonnet 5.5 |
| Agents & tools | 70.6 | 61.9 | Claude Sonnet 5.5 |
| Maths | 71.3 | 67.8 | Claude Sonnet 5.5 |
| Knowledge | 57.4 | 68.5 | Gemini 4 Hydrogen |
| Human preference | – | 68.2 | not comparable |
| Long context | 66.0 | – | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Instruction following, Long context, Multimodal, Human preference.
Where sources disagree
Across the 12 benchmarks both have, Claude Sonnet 5.5 leads 6, Gemini 4 Hydrogen leads 6. Independent sources split: Claude Sonnet 5.5 leads on AA Intelligence Index v5.0 and Vals Index; Gemini 4 Hydrogen leads on Arena Text.
| Overall index | Claude Sonnet 5.5 (max) | Gemini 4 Hydrogen (high) | Leader |
|---|---|---|---|
| BenchLeader Index | 64.7 | 63.4 | neck and neck |
| AA Intelligence Index v5.0 | 54.5 | 50.8 | Claude Sonnet 5.5 |
| Vals Index | 70.8† | 68.5 | Claude Sonnet 5.5 |
| Arena Text | 1473† | 1490 | Gemini 4 Hydrogen |
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. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Claude Sonnet 5.5 (max) | Gemini 4 Hydrogen (high) |
|---|---|---|
| Blended price per 1M tokens | $4.00 | $3.00 |
| Input / output per 1M | $2.00 / $10.00 | $1.50 / $7.50 |
| Cached input per 1M | $0.100 | $0.075 |
| Price per task: chat reply | $0.0038 | $0.0029 |
| Price per task: summarize a 30-page report | $0.030 | $0.022 |
| Price per task: code edit | $0.027 | $0.020 |
| Price per task: agentic coding session | $0.160 | $0.120 |
| Price per task: structured extraction | $0.0060 | $0.0045 |
| Agentic coding session, input mostly cached | $0.074 | $0.056 |
| Cost per benchmark task (AA Index) | $10.02 | $1.87 |
| Output tokens per task (AA Index) | 285k | 85k |
| Of which reasoning | 206k | 48k |
| Time per task (AA Index) | 34 min | 8 min |
| Output speed | 140 tok/s | 180 tok/s |
| Time to first answer | 247.0 s | 16.4 s |
| Time to a complete answer | 487.7 s | 63.3 s |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $10.02 on Claude Sonnet 5.5 (max) and $1.87 on Gemini 4 Hydrogen (high), and Claude Sonnet 5.5 (max) wrote 3.4× as many output tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. Claude Sonnet 5.5 (max) scores 64.7 and takes 247 s to its first answer token; Claude Sonnet 5.5 (xhigh) scores 62.5 in 27 s, Claude Sonnet 5.5 (high) scores 57.9 in 10 s. Gemini 4 Hydrogen (high) scores 63.4 and takes 16 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
117 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 | Claude Sonnet 5.5 (max) | Gemini 4 Hydrogen (high) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 64.7 | 63.4 |
| Agents & tools score | BenchLeader | 70.6 | 61.9 |
| Coding score | BenchLeader | 71.5 | 68.8 |
| Knowledge score | BenchLeader | 57.4 | 68.5 |
| Long context score | BenchLeader | 66.0 | – |
| Maths score | BenchLeader | 71.3 | 67.8 |
| Reasoning score | BenchLeader | 67.7 | 69.6 |
| Human preference score | BenchLeader | – | 68.2 |
| Blended price $/M | BenchLeader | $4.00 | $3.00 |
| Output speed | BenchLeader | 140 tok/s | 180 tok/s |
| Time to first answer | BenchLeader | 247.0 s | 16.4 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 95.6% | – |
| Humanity's Last Exam | Scale AI / CAIS | 46.1%† | – |
| SimpleBench | SimpleBench | 75.9%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1499† | 1514 |
| LiveBench 2.0 Reasoning | LiveBench | 91.6%† | 88.6%† |
| LiveBench 1.0 Reasoning | LiveBench | 91.6% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 55.0% | – |
| HLE-Diamond (AA) | Artificial Analysis | 53.3% | 54.9% |
| ARC-AGI-1 | ARC Prize | 96.5% | – |
| ARC-AGI-2 | ARC Prize | 83.9%† | – |
| ARC-AGI-3 | ARC Prize | 72.4%† | – |
| CritPt | Artificial Analysis | 31.4% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | 59.0% | 61.5% |
| MysteryMechanism | Vals AI | 49.1%† | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 59.0% | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 65.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 68.3% | – |
| DTBench | Epoch AI Benchmarking Hub | 97.3% | – |
| Coding | |||
| SciCode | SciCode | 61.0% | – |
| FrontierCode | Cognition | 52.1%† | – |
| Arena Coding | Arena (LMArena) | 1517† | 1518 |
| Arena WebDev | Arena (LMArena) | 1772† | 1704 |
| LiveBench 2.0 Coding | LiveBench | 77.2%† | 78.5%† |
| LiveBench 1.0 Coding | LiveBench | 91.4% | – |
| IOI | Vals AI | 83.1%† | – |
Show all 117 benchmarks
| Metric | Publisher | Claude Sonnet 5.5 (max) | Gemini 4 Hydrogen (high) |
|---|---|---|---|
| Coding | |||
| Code Migration | Vals AI | 69.8%† | – |
| Code Migration v1.1 | Vals AI | 86.9%† | – |
| ProgramBench | Vals AI | 6.5%† | – |
| SWE-bench Multimodal | Vals AI | 58.3%† | – |
| Vibe Code Bench v1.1 | Vals AI | 92.4%† | 87.8%† |
| MirrorCode | Epoch AI Benchmarking Hub | 97.8% | – |
| CursorBench | Cursor | 55.5% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1819.1† | – |
| FrontierSWE | Epoch AI Benchmarking Hub | 61.9% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 63.6% | 54.0% |
| Startrise Frontend | Startrise | 75.9%† | – |
| GameBench 2 Overall | DuelLab GameBench 2 | 60.3† | – |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 91.2† | – |
| Design Arena Website | Design Arena | 1309† | 1316† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 45.8% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 41.7% | 36.3% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 63.9% | 56.4% |
| APEX-Agents | Mercor | 75.5% | – |
| Arena Agent | Arena (LMArena) | 16 | – |
| LiveBench Long Agentic | LiveBench | 63.4%† | 61.8%† |
| LiveBench Browser | LiveBench | 89.3%† | 77.1%† |
| LiveBench Agentic Coding | LiveBench | 56.3% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 83.2%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 66.9% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1838 | – |
| Analyst Agent (AA) | Artificial Analysis | 57.5% | – |
| CyberBench | Vals AI | 59.6%† | – |
| Finance Agent v2 | Vals AI | 58.1%† | – |
| Finance Agent v2.1 | Vals AI | 67.3%† | 65.1%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 10.4%† | 11.3%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 2.9%† | – |
| Legal Research Bench | Vals AI | 48.1%† | 53.9%† |
| SREBench | Vals AI | 30.1%† | – |
| Tax Agent Bench | Vals AI | 42.0%† | 52.3%† |
| Terminal-Bench 4.0 (Vals) | Vals AI | 64.1%† | 54.0%† |
| Terminal-Bench Science | Vals AI | 47.1%† | – |
| Vending-Bench 2 | Andon Labs | 11165.3† | – |
| AutomationBench | Artificial Analysis | 71.8% | 74.4% |
| GDP.pdf (AA) | Artificial Analysis | 25.8% | 30.0% |
| Harvey LAB-AA v1.1 | Harvey | 2.8% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 1816 | 1497 |
| GDPx | GDPx by SemiAnalysis | 8.2% | – |
| Design Arena Agentic | Design Arena | 1282† | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 53.3% | 41.4% |
| FACTS Search | 86.8%† | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 88.8% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 80.5% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 100.0% | – |
| ProofBench | Vals AI | 100.0% | – |
| LiveBench Mathematics | LiveBench | 96.1% | – |
| Arena Maths | Arena (LMArena) | 1486† | 1495 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 46.5% | – |
| LiveBench 2.0 Data Analysis | LiveBench | 59.5%† | 83.2%† |
| LiveBench Research | LiveBench | 88.0%† | 86.8%† |
| LiveBench 1.0 Data Analysis | LiveBench | 59.5% | – |
| AA-Omniscience v1.1 | Artificial Analysis | 30.5 | 43.7 |
| BioMysteryBench | Vals AI | 81.1%† | – |
| ClaimBench | Vals AI | 100.0%† | – |
| Excel Modeling Benchmark | Vals AI | 75.7%† | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 75.9%† | 75.8%† |
| MedCode | Vals AI | 52.9%† | – |
| MedScribe | Vals AI | 91.1%† | – |
| Public Benefits Bench v1.1 | Vals AI | 67.2%† | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 31.4% | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 51.5% | 48.2% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 56.7% | 91.3% |
| HealthBench Professional (AA) | OpenAI | 63.0% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 78.0% | – |
| LiveBench Instruction Following | LiveBench | 56.8% | – |
| Arena Instruction Following | Arena (LMArena) | 1482† | 1503 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1473† | 1490 |
| Arena Creative Writing | Arena (LMArena) | 1464† | 1492 |
| Arena Multi-turn | Arena (LMArena) | 1469† | 1476 |
| Arena Longer Queries | Arena (LMArena) | 1491† | 1514 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1284† | – |
| LiveBench Design | LiveBench | 81.3%† | 77.3%† |
| SAGE | Vals AI | 88.0%† | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 39.3%† | – |
| Code2Video-AA | Artificial Analysis | 1443 | – |
| AA-3D | Artificial Analysis | 1204 | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 82.7% | – |
| MLCR | Artificial Analysis | 41.7% | 21.1%† |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 166.6† | – |
| LiveBench 2.0 | LiveBench | 77.0%† | 77.0%† |
| LiveBench 1.0 | LiveBench | 75.7% | – |
| AA Intelligence Index v5.0 | Artificial Analysis | 54.5 | 50.8 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 56 | – |
| AA Progress Index v1 | Artificial Analysis | 350.5 | – |
| Vals Index | Vals AI | 70.8† | 68.5† |
| Vals Finance Index | Vals AI | 64.5%† | 66.1%† |
| Vals Science Index | Vals AI | 59.1%† | – |
| Vals Router Index | Vals AI | 69.0%† | – |
Data as of 2026-11-13. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261110-2acac640a228 (https://benchmarks.duellab.org/releases/gb2-20261110-2acac640a228/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- Gemini 4 Hydrogen (high): 10 new results (Arena Text, Arena Hard Prompts, Arena Coding…).
- Claude Sonnet 5.5 (max): BenchLeader Index 65.1 → 64.6.
- Claude Sonnet 5.5 (max): 1 new result (MirrorCode).
- Claude Sonnet 5.5 (max): 2 score revisions.
- Claude Sonnet 5.5 (max): 1 score revision.
- Claude Sonnet 5.5 (max): 1 score revision.
- Claude Sonnet 5.5 (max): 1 score revision.
- Claude Sonnet 5.5 (max): 1 new result (MirrorCode).
- Claude Sonnet 5.5 (max): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; cost per task moved from $5.46/M to $10.02/M (83% up).
- Claude Sonnet 5.5 (max): 1 score revision.
Related comparisons
- Claude Sonnet 5.5 vs Claude Fable 5.2
- Claude Sonnet 5.5 vs Gemini 4 Carbon
- Claude Sonnet 5.5 vs Claude Opus 5.5
- Gemini 4 Hydrogen vs Claude Fable 5.2
- Gemini 4 Hydrogen vs Gemini 4 Carbon
- Gemini 4 Hydrogen vs Claude Opus 5.5
- Gemini 4 Hydrogen vs GPT-6 Astra
- Claude Sonnet 5.5 vs Claude Fable 5.1
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Claude Sonnet 5.5 vs Gemini 4 Hydrogen: questions
- Is Claude Sonnet 5.5 better than Gemini 4 Hydrogen?
- Claude Sonnet 5.5 (max) and Gemini 4 Hydrogen (high) are neck and neck on quality (64.7 vs 63.4, within the ±3.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 13 Nov 2026, so check the category scores for your use.
- Claude Sonnet 5.5 vs Gemini 4 Hydrogen: which is cheaper, and will that change?
- Gemini 4 Hydrogen is cheaper today: $3.00 against $4.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Claude Sonnet 5.5 or Gemini 4 Hydrogen better for coding?
- Claude Sonnet 5.5 scores higher in coding (72 vs 69 on the category index, where 50 is average). On FrontierCode 2.0 (AA), the top shared coding benchmark, it is 41.7% for Claude Sonnet 5.5 against 36.3% for Gemini 4 Hydrogen.
- Which is faster, Claude Sonnet 5.5 or Gemini 4 Hydrogen?
- Gemini 4 Hydrogen streams faster (180 vs 140 output tokens per second, Artificial Analysis 7-day median).
- Is Claude Sonnet 5.5 or Gemini 4 Hydrogen better for agents and tool use?
- Claude Sonnet 5.5 scores higher in agentic tasks (71 vs 62 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 63.9% for Claude Sonnet 5.5 against 56.4% for Gemini 4 Hydrogen.
More questions (4)
- Which is better for long documents?
- Both take about 1M tokens: Claude Sonnet 5.5 1,000,000, Gemini 4 Hydrogen 1,048,576.
- Which has lower latency?
- Gemini 4 Hydrogen answers first: 16.4 s against 247.0 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 Claude Sonnet 5.5 vs Gemini 4 Hydrogen?
- Across the 12 benchmarks both have, Claude Sonnet 5.5 leads 6, Gemini 4 Hydrogen leads 6. Independent sources split: Claude Sonnet 5.5 leads on AA Intelligence Index v5.0 and Vals Index; Gemini 4 Hydrogen leads on Arena Text.
- When was Gemini 4 Hydrogen released?
- Gemini 4 Hydrogen was released on 12 Nov 2026.