Gemini 2.5 Flash vs Hy3
Hy3 costs 7.6x less per output token than Gemini 2.5 Flash ($0.33 vs $2.50 per million). Updated October 21, 2026; 38 benchmarks have a result for both.
Gemini 2.5 Flash or Hy3: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Gemini 2.5 Flash for a larger context window (1.05M against 262K tokens) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
- Shared results
- 38
- Gemini 2.5 Flash only
- 35
- Hy3 only
- 6
- Like-for-like categories
- 6 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
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
No call
Only Gemini 2.5 Flash has a AA Intelligence Index result, 15.5.
Directional only
Web design
Design Arena: websites
Hy3
Hy3 leads, 1230 against 1074.
Same board, both measured
Long documents
Listed context window
Gemini 2.5 Flash
Gemini 2.5 Flash takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3
Gemini 2.5 Flash $0.0015, Hy3 $0.0002 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Gemini 2.5 Flash has the number, $0.466.
Directional only
At a glance
- Coding index
- Gemini 2.5 Flash not benchmarked, Hy3 not benchmarked.
- Widest category
- Web design, 21.9 points apart on average over 7 shared benchmarks.
- Widest single gap
- Design Arena: 3D: Gemini 2.5 Flash 1021 against Hy3 1233, 26.5 points apart on a 0 to 100 scale.
- Output price
- Gemini 2.5 Flash $2.50 against Hy3 $0.33 per million output tokens, 7.6x apart.
- Context window
- Gemini 2.5 Flash takes 1.05M tokens, Hy3 262K.
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 | Gemini 2.5 Flash | Bars | Hy3 | Ties |
|---|---|---|---|---|
| Coding | 0 | 2 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 2 | ||
| Knowledge | 1 | 3 | ||
| Web design | 0 | 7 | ||
| General | 0 | 18 | ||
| All shared boards | 1 | 37 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index02
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| SWE-bench Verified | 28.7% | Not measured | Only Gemini 2.5 Flash measured |
| LMArena WebDev | Not measured | 1507 | Only Hy3 measured |
| LiveCodeBench (Artificial Analysis run) | 69.5% | Not measured | Only Gemini 2.5 Flash measured |
| SciCode | Not measured | 48.6% | Only Hy3 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 13.6% | Not measured | Only Gemini 2.5 Flash measured |
| Terminal-Bench | 17.1% | Not measured | Only Gemini 2.5 Flash measured |
| ALE-Bench | Not measured | 838.6 | Only Hy3 measured |
| LMArena Text: coding | 1423 | 1466 | Hy3 |
| LMArena Text: software and IT | 1425 | 1463 | Hy3 |
| LMArena WebDev: React | Not measured | 1511 | Only Hy3 measured |
| LMArena WebDev: HTML | Not measured | 1487 | Only Hy3 measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 15.5 | Not measured | Only Gemini 2.5 Flash measured |
| Epoch Capabilities Index | 142.9 | Not measured | Only Gemini 2.5 Flash measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 31.6% | Not measured | Only Gemini 2.5 Flash measured |
| BALROG | 33.5% | Not measured | Only Gemini 2.5 Flash measured |
| Vending-Bench 2 | $549 | Not measured | Only Gemini 2.5 Flash measured |
| Berkeley Function Calling Leaderboard (BFCL) | 56.2% | Not measured | Only Gemini 2.5 Flash measured |
| BrowseComp-Plus | 39.9% | Not measured | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 36 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 79.0% | 89.7% | Hy3 |
| Humanity's Last Exam (Artificial Analysis run) | 12.1% | 33.5% | Hy3 |
| ARC-AGI | 33.3% | Not measured | Only Gemini 2.5 Flash measured |
| ARC-AGI-2 | 2.5% | Not measured | Only Gemini 2.5 Flash measured |
| SimpleBench | 41.2% | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Text: hard prompts | 1422 | 1447 | Hy3 |
| LMArena Text: expert | 1426 | 1461 | Hy3 |
| LMArena Text: hard prompts in English | 1420 | 1455 | Hy3 |
| LMArena Vision: diagrams | 1245 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: homework | 1244 | Not measured | Only Gemini 2.5 Flash measured |
| Kagi LLM Benchmark | 56.8% | Not measured | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 11 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| AA Math Index | 73.3 | Not measured | Only Gemini 2.5 Flash measured |
| AIME 2025 (Artificial Analysis run) | 73.3% | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Text: math | 1411 | 1473 | Hy3 |
| LMArena Text: mathematical work | 1422 | 1482 | Hy3 |
| Neither model has a result on the other 9 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| MedCode (Vals AI) | 40.4% | Not measured | Only Gemini 2.5 Flash measured |
| MMLU Pro (Artificial Analysis run) | 83.2% | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Text: science | 1433 | 1457 | Hy3 |
| LMArena Text: medicine and healthcare | 1428 | 1452 | Hy3 |
| LMArena Text: legal and government | 1430 | 1422 | Gemini 2.5 Flash |
| LMArena Text: business and finance | 1404 | 1424 | Hy3 |
| FACTS Benchmark Suite | 47.2% | Not measured | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 11 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index07
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1074 | 1230 | Hy3 |
| Design Arena: UI components | 1005 | 1207 | Hy3 |
| Design Arena: game dev | 1005 | 1208 | Hy3 |
| Design Arena: data viz | 1060 | 1163 | Hy3 |
| Design Arena: 3D | 1021 | 1233 | Hy3 |
| Design Arena: all code categories | 1059 | 1227 | Hy3 |
| Design Arena: SVG | 1020 | 1200 | Hy3 |
| Design Arena: ASCII art | Not measured | 1152 | Only Hy3 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
| Benchmark | Gemini 2.5 Flash | Hy3 | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 44.8% | Not measured | Only Gemini 2.5 Flash measured |
| MedScribe (Vals AI) | 83.0% | Not measured | Only Gemini 2.5 Flash measured |
| IFBench | 50.3% | Not measured | Only Gemini 2.5 Flash measured |
| Long context reasoning (AA-LCR) | 65.3% | 79.0% | Hy3 |
| DTBench | 76.5% | Not measured | Only Gemini 2.5 Flash measured |
| LMCA | 27.5% | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Text | 1417 | 1438 | Hy3 |
| LMArena Text: instruction following | 1405 | 1426 | Hy3 |
| LMArena Text: creative writing | 1400 | 1403 | Hy3 |
| LMArena Vision | 1235 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Text: English | 1418 | 1448 | Hy3 |
| LMArena Text: non-English | 1409 | 1426 | Hy3 |
| LMArena Text: Chinese | 1449 | 1485 | Hy3 |
| LMArena Text: German | 1417 | 1449 | Hy3 |
| LMArena Text: Japanese | 1396 | 1397 | Hy3 |
| LMArena Text: Korean | 1385 | 1396 | Hy3 |
| LMArena Text: French | 1432 | 1456 | Hy3 |
| LMArena Text: Spanish | 1422 | 1457 | Hy3 |
| LMArena Text: Russian | 1415 | 1432 | Hy3 |
| LMArena Text: Polish | 1413 | 1447 | Hy3 |
| LMArena Text: multi-turn | 1408 | 1436 | Hy3 |
| LMArena Text: longer queries | 1419 | 1442 | Hy3 |
| LMArena Text: writing and language | 1405 | 1414 | Hy3 |
| LMArena Text: entertainment and media | 1393 | 1400 | Hy3 |
| LMArena Vision: English | 1239 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: Chinese | 1284 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: reading text | 1237 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: humor | 1225 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: captioning | 1277 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: recognizing things | 1252 | Not measured | Only Gemini 2.5 Flash measured |
| LMArena Vision: creative writing | 1241 | Not measured | Only Gemini 2.5 Flash measured |
| Neither model has a result on the other 15 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemini 2.5 Flash | Hy3 | 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
Hy3 costs less
Repository review
50K input + 3K output tokens
Hy3 costs less
Agent loop
200K input + 10K output tokens
Hy3 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 | Google Gemini 2.5 Flash | Tencent Hy3 |
|---|---|---|
| Maker | Tencent | |
| Weights | No public weights listed | Open weights |
| Released | June 17, 2025 | July 6, 2026 |
| Context window | 1.05M | 262K |
| Input $/Mtok | $0.30 | $0.0825 |
| Output $/Mtok | $2.50 | $0.33 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 1 |
| Cost per solved task | $0.466 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemini 2.5 Flash better than Hy3 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Gemini 2.5 Flash for a larger context window (1.05M against 262K tokens) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
Which is cheaper, Gemini 2.5 Flash or Hy3?
Gemini 2.5 Flash costs $0.30 per million input tokens and $2.50 per million output tokens; Hy3 costs $0.0825 and $0.33. Cost per solved task is known for Gemini 2.5 Flash only.
Which has the bigger context window, Gemini 2.5 Flash or Hy3?
Gemini 2.5 Flash, with 1.05M tokens against 262K.
How many benchmarks is the Gemini 2.5 Flash vs Hy3 comparison based on?
38 benchmarks have a result for both models as of October 21, 2026. Gemini 2.5 Flash's coding index rests on 0 independent benchmarks and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.