LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Claude Haiku 5.5 vs Gemini 4 Argon

Updated October 21, 2026. 62 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Claude Haiku 5.5 Replace Gemini 4 Argon
Claude Haiku 5.5Anthropic
Coding
75.9
Overall
43.4
Output $/M
$0.50
4
Gemini 4 ArgonGoogle
Coding
80.5
Overall
52.6
Output $/M
$10.00
57
vs boards won of 62 shared, 1 tied

Decision reading

Pick Gemini 4 Argon for coding: it scores 80.5 on the Vibe Coding Index against 75.9 for Claude Haiku 5.5.

Pick Claude Haiku 5.5 when output tokens cost 20.0x less, or when what matters is LiveBench coding or LiveBench design, where it leads.

Model AClaude Haiku 5.5Anthropic75.9/1002 independent benchmarksNo cost per solved taskModel BGemini 4 ArgonGoogle80.5/1002 independent benchmarksNo cost per solved task
Shared results
62
Claude Haiku 5.5 only
0
Gemini 4 Argon only
30
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Claude Haiku 5.5 75.9 from 2 benchmarks, Gemini 4 Argon 80.5 from 2 benchmarks.
Widest category
Knowledge, 29.7 points apart on average over 2 shared benchmarks.
Widest single gap
Vending-Bench 2: Claude Haiku 5.5 $3,602 against Gemini 4 Argon $13,718, 63.2 points apart on a 0 to 100 scale.
Output price
Claude Haiku 5.5 $0.50 against Gemini 4 Argon $10.00 per million output tokens, 20.0x apart.
Context window
Claude Haiku 5.5 takes 1M tokens, Gemini 4 Argon 1.05M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryClaude Haiku 5.5BarsGemini 4 ArgonTies
Coding1131 tie
Intelligence02
Agentic work014
Reasoning09
Math04
Knowledge02
Web design16
General27
All shared boards4571 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index113
Coding benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
LiveBench coding91.4%89.7%Claude Haiku 5.5
LMArena WebDev15861678Gemini 4 Argon
Terminal-Bench 4 (Vals AI)36.4%57.6%Gemini 4 Argon
Vibe Code Bench (Vals AI)90.4%91.9%Gemini 4 Argon
Code migration (Vals AI)44.6%68.2%Gemini 4 Argon
ProgramBench (Vals AI)2.5%2.5%Tie
IOI (Vals AI)47.3%100.0%Gemini 4 Argon
Terminal-Bench 2.1 (Vals AI)Not measured57.7%Only Gemini 4 Argon measured
SWE-bench Multimodal (Vals AI)44.0%56.0%Gemini 4 Argon
Vibe Code Bench 1-100 (Vals AI)20.4%21.2%Gemini 4 Argon
CyberBench (Vals AI)75.5%77.9%Gemini 4 Argon
SciCode55.0%61.8%Gemini 4 Argon
FrontierCode46.4%47.5%Gemini 4 Argon
FrontierSWE43.8%55.0%Gemini 4 Argon
ALE-BenchNot measured2100.1Only Gemini 4 Argon measured
LMArena Text: codingNot measured1546Only Gemini 4 Argon measured
LMArena Text: software and ITNot measured1556Only Gemini 4 Argon measured
LMArena WebDev: React16051692Gemini 4 Argon
LMArena WebDev: HTML15691643Gemini 4 Argon
Neither model has a result on the other 19 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
AA Intelligence Index43.452.6Gemini 4 Argon
Epoch Capabilities Index159.3166.4Gemini 4 Argon
Agentic workLike-for-likePublic boards, shown as published, kept out of the index014
Agentic work benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
Design Arena: spreadsheetsNot measured1151Only Gemini 4 Argon measured
LiveBench research81.9%83.1%Gemini 4 Argon
LiveBench browser80.0%92.0%Gemini 4 Argon
LiveBench long agentic57.6%69.9%Gemini 4 Argon
SREBench (Vals AI)5.7%44.3%Gemini 4 Argon
Finance Agent (Vals AI)54.1%65.4%Gemini 4 Argon
Tax Agent Bench (Vals AI)31.9%44.6%Gemini 4 Argon
Legal Research Bench (Vals AI)43.3%54.8%Gemini 4 Argon
Harvey's Legal Agent Benchmark (Vals AI)4.2%27.1%Gemini 4 Argon
Excel Modeling Benchmark (Vals AI)59.0%75.2%Gemini 4 Argon
BioMysteryBench (Vals AI)70.4%76.3%Gemini 4 Argon
Terminal-Bench-Science (Vals AI)10.0%44.3%Gemini 4 Argon
Vals Science Index38.5%55.4%Gemini 4 Argon
Time Horizon Index (Vals AI)Not measured52.5%Only Gemini 4 Argon measured
CUA-bench (Vals AI)Not measured4.8%Only Gemini 4 Argon measured
Robo-bench (Vals AI)Not measured0.8%Only Gemini 4 Argon measured
APEX-Agents56.1%82.4%Gemini 4 Argon
Vending-Bench 2$3,602$13,718Gemini 4 Argon
Neither model has a result on the other 21 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index09
Reasoning benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
LiveBench reasoning82.4%95.1%Gemini 4 Argon
MysteryMechanism (Vals AI)35.1%45.5%Gemini 4 Argon
Humanity's Last Exam (Artificial Analysis run)44.4%57.1%Gemini 4 Argon
GPQA Diamond (Epoch AI run)89.6%96.4%Gemini 4 Argon
ARC-AGI89.0%98.5%Gemini 4 Argon
ARC-AGI-261.3%91.3%Gemini 4 Argon
Chess puzzles55.0%59.0%Gemini 4 Argon
Mystery game puzzles30.0%64.0%Gemini 4 Argon
EBR-bench43.8%64.8%Gemini 4 Argon
LMArena Text: hard promptsNot measured1546Only Gemini 4 Argon measured
LMArena Text: expertNot measured1550Only Gemini 4 Argon measured
LMArena Text: hard prompts in EnglishNot measured1548Only Gemini 4 Argon measured
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index04
Math benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
OTIS Mock AIME98.9%100.0%Gemini 4 Argon
FrontierMath, tiers 1 to 375.1%86.0%Gemini 4 Argon
FrontierMath, tier 446.3%78.0%Gemini 4 Argon
ProofBench89.0%99.0%Gemini 4 Argon
LMArena Text: mathNot measured1536Only Gemini 4 Argon measured
LMArena Text: mathematical workNot measured1557Only Gemini 4 Argon measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index02
Knowledge benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
LegalBench (Vals AI)Not measured88.3%Only Gemini 4 Argon measured
MedCode (Vals AI)51.5%58.8%Gemini 4 Argon
SimpleQA Verified23.8%76.0%Gemini 4 Argon
LMArena Text: scienceNot measured1548Only Gemini 4 Argon measured
LMArena Text: medicine and healthcareNot measured1555Only Gemini 4 Argon measured
LMArena Text: legal and governmentNot measured1548Only Gemini 4 Argon measured
LMArena Text: business and financeNot measured1525Only Gemini 4 Argon measured
Neither model has a result on the other 10 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index16
Web design benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
Design Arena: websites12631315Gemini 4 Argon
Design Arena: UI components12951333Gemini 4 Argon
Design Arena: game dev13151377Gemini 4 Argon
Design Arena: data viz12921334Gemini 4 Argon
Design Arena: 3D13431389Gemini 4 Argon
Design Arena: all code categories12791328Gemini 4 Argon
LiveBench design83.1%73.7%Claude Haiku 5.5
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index27
General benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
Vals Index58.7%69.2%Gemini 4 Argon
LiveBench data analysis53.1%85.2%Gemini 4 Argon
Public Benefits Bench (Vals AI)63.2%69.8%Gemini 4 Argon
SAGE (Vals AI)45.8%53.6%Gemini 4 Argon
MedScribe (Vals AI)89.5%87.4%Claude Haiku 5.5
Long context reasoning (AA-LCR)82.7%79.7%Claude Haiku 5.5
DTBench95.0%97.6%Gemini 4 Argon
LMCA51.1%63.7%Gemini 4 Argon
Furniture assembly47.5%72.5%Gemini 4 Argon
LMArena TextNot measured1532Only Gemini 4 Argon measured
LMArena Text: instruction followingNot measured1534Only Gemini 4 Argon measured
LMArena Text: creative writingNot measured1527Only Gemini 4 Argon measured
LMArena Text: EnglishNot measured1538Only Gemini 4 Argon measured
LMArena Text: non-EnglishNot measured1522Only Gemini 4 Argon measured
LMArena Text: ChineseNot measured1611Only Gemini 4 Argon measured
LMArena Text: SpanishNot measured1503Only Gemini 4 Argon measured
LMArena Text: RussianNot measured1534Only Gemini 4 Argon measured
LMArena Text: multi-turnNot measured1556Only Gemini 4 Argon measured
LMArena Text: longer queriesNot measured1547Only Gemini 4 Argon measured
LMArena Text: writing and languageNot measured1529Only Gemini 4 Argon measured
LMArena Text: entertainment and mediaNot measured1527Only Gemini 4 Argon measured
Neither model has a result on the other 25 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Haiku 5.5 and Gemini 4 Argon
BenchmarkClaude Haiku 5.5Gemini 4 ArgonLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Haiku 5.5$0.0003Fits in one request
Gemini 4 Argon$0.0070Fits in one request

Claude Haiku 5.5 costs less

Repository review

50K input + 3K output tokens

Claude Haiku 5.5$0.0065Fits in one request
Gemini 4 Argon$0.130Fits in one request

Claude Haiku 5.5 costs less

Agent loop

200K input + 10K output tokens

Claude Haiku 5.5$0.025Fits in one request
Gemini 4 Argon$0.500Fits in one request

Claude Haiku 5.5 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Haiku 5.5 and Gemini 4 Argon
SpecificationAnthropic Claude Haiku 5.5Google Gemini 4 Argon
MakerAnthropicGoogle
WeightsNo public weights listedNo public weights listed
ReleasedOctober 7, 2026October 13, 2026
Context window1M1.05M
Input $/Mtok$0.10$2.00
Output $/Mtok$0.50$10.00
Coding index75.980.5
Independent measurements behind it22
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Haiku 5.5 better than Gemini 4 Argon for coding?

Pick Gemini 4 Argon for coding: it scores 80.5 on the Vibe Coding Index against 75.9 for Claude Haiku 5.5. Pick Claude Haiku 5.5 when output tokens cost 20.0x less, or when what matters is LiveBench coding or LiveBench design, where it leads.

Which is cheaper, Claude Haiku 5.5 or Gemini 4 Argon?

Claude Haiku 5.5 costs $0.10 per million input tokens and $0.50 per million output tokens; Gemini 4 Argon costs $2.00 and $10.00. Cost per solved task is known for neither of them.

Which has the bigger context window, Claude Haiku 5.5 or Gemini 4 Argon?

Gemini 4 Argon, with 1.05M tokens against 1M.

How many benchmarks is the Claude Haiku 5.5 vs Gemini 4 Argon comparison based on?

62 benchmarks have a result for both models as of October 21, 2026. Claude Haiku 5.5's coding index rests on 2 independent benchmarks and Gemini 4 Argon's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/claude-haiku-5-5-vs-gemini-4-argon" title="Claude Haiku 5.5 vs Gemini 4 Argon, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/claude-haiku-5-5-vs-gemini-4-argon">LLMBenchmark.ai, Claude Haiku 5.5 vs Gemini 4 Argon</a></p>