Claude Haiku 4.5 vs Gemini 2.5 Flash
Claude Haiku 4.5 leads Gemini 2.5 Flash on the Artificial Analysis Intelligence Index, 16.9 to 15.5, and Gemini 2.5 Flash costs 2.0x less per output token ($2.50 vs $5.00 per million). Updated October 21, 2026; 49 benchmarks have a result for both.
Claude Haiku 4.5 or Gemini 2.5 Flash: which to pick
Only Claude Haiku 4.5 has a coding score, 35.6 on the Vibe Coding Index; Gemini 2.5 Flash has none, so the two cannot be ranked on coding.
Pick Gemini 2.5 Flash for output tokens 2.0x cheaper ($2.50 vs $5.00 per million) and a lower cost per solved task ($0.466 against $0.56).
- Shared results
- 49
- Claude Haiku 4.5 only
- 46
- Gemini 2.5 Flash only
- 24
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
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.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Only Claude Haiku 4.5 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Claude Haiku 4.5
Claude Haiku 4.5 leads, 16.9 against 15.5.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Claude Haiku 4.5 has a AA Agentic Index result, 8.0.
Directional only
Reasoning
GPQA Diamond
No call
Only Claude Haiku 4.5 has a GPQA Diamond result, 72.2%.
Directional only
Math
AIME
No call
Only Claude Haiku 4.5 has a AIME result, 82.7%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Haiku 4.5 has a MMLU Pro result, 78.7%.
Directional only
Web design
Design Arena: websites
Claude Haiku 4.5
Claude Haiku 4.5 leads, 1122 against 1074.
Same board, both measured
Long documents
Listed context window
Gemini 2.5 Flash
Gemini 2.5 Flash takes 1.05M tokens against 200K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Gemini 2.5 Flash
Claude Haiku 4.5 $0.0035, Gemini 2.5 Flash $0.0015 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Gemini 2.5 Flash
Claude Haiku 4.5 $0.56, Gemini 2.5 Flash $0.466.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Haiku 4.5 35.6 from 4 benchmarks, Gemini 2.5 Flash not benchmarked.
- Widest category
- Coding, 15.9 points apart on average over 4 shared benchmarks.
- Widest single gap
- SWE-bench Verified: Claude Haiku 4.5 66.6% against Gemini 2.5 Flash 28.7%, 37.9 points apart on a 0 to 100 scale.
- Output price
- Claude Haiku 4.5 $5.00 against Gemini 2.5 Flash $2.50 per million output tokens, 2.0x apart.
- Context window
- Claude Haiku 4.5 takes 200K tokens, Gemini 2.5 Flash 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.
| Category | Claude Haiku 4.5 | Bars | Gemini 2.5 Flash | Ties |
|---|---|---|---|---|
| Coding | 4 | 0 | ||
| Intelligence | 1 | 1 | ||
| Agentic work | 1 | 2 | ||
| Reasoning | 4 | 1 | ||
| Math | 1 | 1 | ||
| Knowledge | 0 | 5 | ||
| Web design | 7 | 0 | ||
| General | 5 | 16 | ||
| All shared boards | 23 | 26 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index40
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| SWE-bench Verified | 66.6% | 28.7% | Claude Haiku 4.5 |
| SWE-bench Multilingual | 64.7% | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench coding | 48.1% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena WebDev | 1330 | Not measured | Only Claude Haiku 4.5 measured |
| AA Coding Index | 43.9 | Not measured | Only Claude Haiku 4.5 measured |
| LiveCodeBench (Vals AI) | 41.2% | Not measured | Only Claude Haiku 4.5 measured |
| SWE-bench Verified (Vals AI) | 66.6% | Not measured | Only Claude Haiku 4.5 measured |
| Vibe Code Bench (Vals AI) | 11.4% | Not measured | Only Claude Haiku 4.5 measured |
| Code migration (Vals AI) | 7.5% | Not measured | Only Claude Haiku 4.5 measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only Claude Haiku 4.5 measured |
| Terminal-Bench 2.1 (Vals AI) | 43.8% | Not measured | Only Claude Haiku 4.5 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 69.5% | Only Gemini 2.5 Flash measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 13.6% | Only Gemini 2.5 Flash measured |
| Terminal-Bench | 35.5% | 17.1% | Claude Haiku 4.5 |
| WeirdML | 45.4% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: coding | 1454 | 1423 | Claude Haiku 4.5 |
| LMArena Text: software and IT | 1439 | 1425 | Claude Haiku 4.5 |
| LMArena WebDev: React | 1326 | Not measured | Only Claude Haiku 4.5 measured |
| LMArena WebDev: HTML | 1318 | Not measured | Only Claude Haiku 4.5 measured |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| AA Intelligence Index | 16.9 | 15.5 | Claude Haiku 4.5 |
| Epoch Capabilities Index | 142.2 | 142.9 | Gemini 2.5 Flash |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| AA Agentic Index | 8.0 | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench research | 42.2% | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench browser | 40.0% | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench long agentic | 11.1% | Not measured | Only Claude Haiku 4.5 measured |
| Finance Agent (Vals AI) | 31.0% | Not measured | Only Claude Haiku 4.5 measured |
| Tax Agent Bench (Vals AI) | 5.9% | Not measured | Only Claude Haiku 4.5 measured |
| Legal Research Bench (Vals AI) | 10.6% | Not measured | Only Claude Haiku 4.5 measured |
| Excel Modeling Benchmark (Vals AI) | 22.7% | Not measured | Only Claude Haiku 4.5 measured |
| τ²-Bench Telecom | Not measured | 31.6% | Only Gemini 2.5 Flash measured |
| BALROG | 31.2% | 33.5% | Gemini 2.5 Flash |
| DeepResearch Bench | 45.5% | Not measured | Only Claude Haiku 4.5 measured |
| Vending-Bench 2 | $459 | $549 | Gemini 2.5 Flash |
| Berkeley Function Calling Leaderboard (BFCL) | 68.7% | 56.2% | Claude Haiku 4.5 |
| BrowseComp-Plus | Not measured | 39.9% | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index41
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| GPQA Diamond | 72.2% | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench reasoning | 61.7% | Not measured | Only Claude Haiku 4.5 measured |
| MMMU Pro (Vals AI) | 46.1% | Not measured | Only Claude Haiku 4.5 measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 79.0% | Only Gemini 2.5 Flash measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 12.1% | Only Gemini 2.5 Flash measured |
| GPQA Diamond (Epoch AI run) | 71.2% | Not measured | Only Claude Haiku 4.5 measured |
| ARC-AGI | 47.7% | 33.3% | Claude Haiku 4.5 |
| ARC-AGI-2 | 4.0% | 2.5% | Claude Haiku 4.5 |
| SimpleBench | Not measured | 41.2% | Only Gemini 2.5 Flash measured |
| Chess puzzles | 8.0% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: hard prompts | 1421 | 1422 | Gemini 2.5 Flash |
| LMArena Text: expert | 1443 | 1426 | Claude Haiku 4.5 |
| LMArena Text: hard prompts in English | 1435 | 1420 | Claude Haiku 4.5 |
| LMArena Vision: diagrams | Not measured | 1245 | Only Gemini 2.5 Flash measured |
| LMArena Vision: homework | Not measured | 1244 | Only Gemini 2.5 Flash measured |
| Kagi LLM Benchmark | Not measured | 56.8% | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 6 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| AIME | 82.7% | Not measured | Only Claude Haiku 4.5 measured |
| MGSM (Vals AI) | 92.2% | Not measured | Only Claude Haiku 4.5 measured |
| AA Math Index | Not measured | 73.3 | Only Gemini 2.5 Flash measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 73.3% | Only Gemini 2.5 Flash measured |
| OTIS Mock AIME | 66.7% | Not measured | Only Claude Haiku 4.5 measured |
| MATH level 5 | 96.4% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: math | 1398 | 1411 | Gemini 2.5 Flash |
| LMArena Text: mathematical work | 1430 | 1422 | Claude Haiku 4.5 |
| Neither model has a result on the other 5 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| MMLU Pro | 78.7% | Not measured | Only Claude Haiku 4.5 measured |
| LegalBench (Vals AI) | 81.2% | Not measured | Only Claude Haiku 4.5 measured |
| CaseLaw (Vals AI) | 56.5% | Not measured | Only Claude Haiku 4.5 measured |
| CorpFin (Vals AI) | 60.6% | Not measured | Only Claude Haiku 4.5 measured |
| TaxEval (Vals AI) | 67.5% | Not measured | Only Claude Haiku 4.5 measured |
| MortgageTax (Vals AI) | 62.2% | Not measured | Only Claude Haiku 4.5 measured |
| MedQA (Vals AI) | 79.6% | Not measured | Only Claude Haiku 4.5 measured |
| MedCode (Vals AI) | 32.7% | 40.4% | Gemini 2.5 Flash |
| MMLU Pro (Artificial Analysis run) | Not measured | 83.2% | Only Gemini 2.5 Flash measured |
| SimpleQA Verified | 13.2% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text: science | 1406 | 1433 | Gemini 2.5 Flash |
| LMArena Text: medicine and healthcare | 1394 | 1428 | Gemini 2.5 Flash |
| LMArena Text: legal and government | 1400 | 1430 | Gemini 2.5 Flash |
| LMArena Text: business and finance | 1400 | 1404 | Gemini 2.5 Flash |
| FACTS Benchmark Suite | Not measured | 47.2% | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 3 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index70
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| Design Arena: websites | 1122 | 1074 | Claude Haiku 4.5 |
| Design Arena: UI components | 1100 | 1005 | Claude Haiku 4.5 |
| Design Arena: game dev | 1097 | 1005 | Claude Haiku 4.5 |
| Design Arena: data viz | 1124 | 1060 | Claude Haiku 4.5 |
| Design Arena: 3D | 1072 | 1021 | Claude Haiku 4.5 |
| Design Arena: all code categories | 1115 | 1059 | Claude Haiku 4.5 |
| Design Arena: SVG | 1026 | 1020 | Claude Haiku 4.5 |
| Design Arena: ASCII art | 1140 | Not measured | Only Claude Haiku 4.5 measured |
| LiveBench design | 59.9% | Not measured | Only Claude Haiku 4.5 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index516
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| LiveBench data analysis | 59.3% | Not measured | Only Claude Haiku 4.5 measured |
| Public Benefits Bench (Vals AI) | 54.3% | Not measured | Only Claude Haiku 4.5 measured |
| SAGE (Vals AI) | 31.8% | 44.8% | Gemini 2.5 Flash |
| MedScribe (Vals AI) | 85.2% | 83.0% | Claude Haiku 4.5 |
| Vals Multimodal Index | 42.9% | Not measured | Only Claude Haiku 4.5 measured |
| IFBench | Not measured | 50.3% | Only Gemini 2.5 Flash measured |
| Long context reasoning (AA-LCR) | Not measured | 65.3% | Only Gemini 2.5 Flash measured |
| DTBench | 73.6% | 76.5% | Gemini 2.5 Flash |
| LMCA | 30.9% | 27.5% | Claude Haiku 4.5 |
| ExploitBench | 13.7% | Not measured | Only Claude Haiku 4.5 measured |
| LMArena Text | 1397 | 1417 | Gemini 2.5 Flash |
| LMArena Text: instruction following | 1414 | 1405 | Claude Haiku 4.5 |
| LMArena Text: creative writing | 1371 | 1400 | Gemini 2.5 Flash |
| LMArena Vision | Not measured | 1235 | Only Gemini 2.5 Flash measured |
| LMArena Text: English | 1414 | 1418 | Gemini 2.5 Flash |
| LMArena Text: non-English | 1377 | 1409 | Gemini 2.5 Flash |
| LMArena Text: Chinese | 1418 | 1449 | Gemini 2.5 Flash |
| LMArena Text: German | 1374 | 1417 | Gemini 2.5 Flash |
| LMArena Text: Japanese | 1338 | 1396 | Gemini 2.5 Flash |
| LMArena Text: Korean | 1346 | 1385 | Gemini 2.5 Flash |
| LMArena Text: French | 1411 | 1432 | Gemini 2.5 Flash |
| LMArena Text: Spanish | 1421 | 1422 | Gemini 2.5 Flash |
| LMArena Text: Russian | 1381 | 1415 | Gemini 2.5 Flash |
| LMArena Text: Polish | 1343 | 1413 | Gemini 2.5 Flash |
| LMArena Text: multi-turn | 1409 | 1408 | Claude Haiku 4.5 |
| LMArena Text: longer queries | 1427 | 1419 | Claude Haiku 4.5 |
| LMArena Text: writing and language | 1386 | 1405 | Gemini 2.5 Flash |
| LMArena Text: entertainment and media | 1369 | 1393 | Gemini 2.5 Flash |
| LMArena Vision: English | Not measured | 1239 | Only Gemini 2.5 Flash measured |
| LMArena Vision: Chinese | Not measured | 1284 | Only Gemini 2.5 Flash measured |
| LMArena Vision: reading text | Not measured | 1237 | Only Gemini 2.5 Flash measured |
| LMArena Vision: humor | Not measured | 1225 | Only Gemini 2.5 Flash measured |
| LMArena Vision: captioning | Not measured | 1277 | Only Gemini 2.5 Flash measured |
| LMArena Vision: recognizing things | Not measured | 1252 | Only Gemini 2.5 Flash measured |
| LMArena Vision: creative writing | Not measured | 1241 | Only Gemini 2.5 Flash measured |
| LMArena Document | 1420 | Not measured | Only Claude Haiku 4.5 measured |
| Neither model has a result on the other 10 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Haiku 4.5 | Gemini 2.5 Flash | Leader |
|---|---|---|---|
| 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
Gemini 2.5 Flash costs less
Repository review
50K input + 3K output tokens
Gemini 2.5 Flash costs less
Agent loop
200K input + 10K output tokens
Gemini 2.5 Flash 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.
| Specification | Anthropic Claude Haiku 4.5 | Google Gemini 2.5 Flash |
|---|---|---|
| Maker | Anthropic | |
| Weights | No public weights listed | No public weights listed |
| Released | October 15, 2025 | June 17, 2025 |
| Context window | 200K | 1.05M |
| Input $/Mtok | $1.00 | $0.30 |
| Output $/Mtok | $5.00 | $2.50 |
| Coding index | 35.6 | Not benchmarked |
| Independent measurements behind it | 4 | 0 |
| Cost per solved task | $0.56 | $0.466 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Haiku 4.5 better than Gemini 2.5 Flash for coding?
Only Claude Haiku 4.5 has a coding score, 35.6 on the Vibe Coding Index; Gemini 2.5 Flash has none, so the two cannot be ranked on coding. Pick Gemini 2.5 Flash for output tokens 2.0x cheaper ($2.50 vs $5.00 per million) and a lower cost per solved task ($0.466 against $0.56).
Which is cheaper, Claude Haiku 4.5 or Gemini 2.5 Flash?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Gemini 2.5 Flash costs $0.30 and $2.50. Per solved SWE-bench Verified task, Claude Haiku 4.5 costs $0.56 and Gemini 2.5 Flash $0.466.
Which has the bigger context window, Claude Haiku 4.5 or Gemini 2.5 Flash?
Gemini 2.5 Flash, with 1.05M tokens against 200K.
How many benchmarks is the Claude Haiku 4.5 vs Gemini 2.5 Flash comparison based on?
49 benchmarks have a result for both models as of October 21, 2026. Claude Haiku 4.5's coding index rests on 4 independent benchmarks and Gemini 2.5 Flash's on 0. Where only one model has a result, the page says so instead of naming a winner.