Qwen3 30B A3B vs Reka Flash 3
Qwen3 30B A3B leads Reka Flash 3 on the Artificial Analysis Intelligence Index, 7.6 to 6.4, and Reka Flash 3 costs 2.5x less per output token ($0.20 vs $0.50 per million). Updated October 23, 2026; 11 benchmarks have a result for both.
Qwen3 30B A3B or Reka Flash 3: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Compare them on price, context and the boards below.
- Shared results
- 11
- Qwen3 30B A3B only
- 42
- Reka Flash 3 only
- 1
- Like-for-like categories
- 7 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
Qwen3 30B A3B
Qwen3 30B A3B leads, 7.6 against 6.4.
Same board, both measured
Web design
Design Arena: websites
No call
Only Qwen3 30B A3B has a Design Arena: websites result, 954.
Directional only
Long documents
Listed context window
Reka Flash 3
Reka Flash 3 takes 66K tokens against 41K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Reka Flash 3
Qwen3 30B A3B $0.0004, Reka Flash 3 $0.0002 per turn.
List prices
At a glance
- Coding index
- Qwen3 30B A3B not benchmarked, Reka Flash 3 not benchmarked.
- Widest category
- Math, 38.6 points apart on average over 2 shared benchmarks.
- Widest single gap
- AIME 2025 (Artificial Analysis run): Qwen3 30B A3B 72.3% against Reka Flash 3 33.7%, 38.7 points apart on a 0 to 100 scale.
- Output price
- Qwen3 30B A3B $0.50 against Reka Flash 3 $0.20 per million output tokens, 2.5x apart.
- Context window
- Qwen3 30B A3B takes 41K tokens, Reka Flash 3 66K.
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 | Qwen3 30B A3B | Bars | Reka Flash 3 | Ties |
|---|---|---|---|---|
| Coding | 2 | 0 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 2 | 0 | ||
| Math | 2 | 0 | ||
| Knowledge | 1 | 0 | ||
| General | 1 | 0 | 1 tie | |
| All shared boards | 10 | 0 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| LiveCodeBench (Artificial Analysis run) | 50.6% | 43.5% | Qwen3 30B A3B |
| Terminal-Bench Hard (Artificial Analysis run) | 6.8% | 0.0% | Qwen3 30B A3B |
| WeirdML | 29.8% | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: coding | 1337 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: software and IT | 1340 | Not measured | Only Qwen3 30B A3B measured |
| Neither model has a result on the other 38 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 7.6 | 6.4 | Qwen3 30B A3B |
| Epoch Capabilities Index | 137.3 | Not measured | Only Qwen3 30B A3B measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 26.0% | 0.0% | Qwen3 30B A3B |
| BALROG | Not measured | 29.2% | Only Reka Flash 3 measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 61.6% | 52.9% | Qwen3 30B A3B |
| Humanity's Last Exam (Artificial Analysis run) | 6.2% | 4.4% | Qwen3 30B A3B |
| GPQA Diamond (Epoch AI run) | 61.7% | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: hard prompts | 1314 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: expert | 1314 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: hard prompts in English | 1329 | Not measured | Only Qwen3 30B A3B measured |
| Neither model has a result on the other 17 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| AA Math Index | 72.3 | 33.7 | Qwen3 30B A3B |
| AIME 2025 (Artificial Analysis run) | 72.3% | 33.7% | Qwen3 30B A3B |
| OTIS Mock AIME | 62.8% | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: math | 1355 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: mathematical work | 1363 | Not measured | Only Qwen3 30B A3B measured |
| MathArena | 9.4% | Not measured | Only Qwen3 30B A3B measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 77.7% | 66.9% | Qwen3 30B A3B |
| LMArena Text: science | 1321 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: medicine and healthcare | 1318 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: legal and government | 1314 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: business and finance | 1308 | Not measured | Only Qwen3 30B A3B measured |
| Neither model has a result on the other 15 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| Design Arena: websites | 954 | Not measured | Only Qwen3 30B A3B measured |
| Design Arena: UI components | 943 | Not measured | Only Qwen3 30B A3B measured |
| Design Arena: game dev | 898 | Not measured | Only Qwen3 30B A3B measured |
| Design Arena: data viz | 967 | Not measured | Only Qwen3 30B A3B measured |
| Design Arena: all code categories | 941 | Not measured | Only Qwen3 30B A3B measured |
| Neither model has a result on the other 4 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | Leader |
|---|---|---|---|
| IFBench | 41.5% | 30.4% | Qwen3 30B A3B |
| Long context reasoning (AA-LCR) | 0.0% | 0.0% | Tie |
| DTBench | 60.3% | Not measured | Only Qwen3 30B A3B measured |
| LMCA | 15.8% | Not measured | Only Qwen3 30B A3B measured |
| Fiction.LiveBench | 40.6% | Not measured | Only Qwen3 30B A3B measured |
| Creative writing (Lech Mazur) | 75.3% | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text | 1317 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: instruction following | 1283 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: creative writing | 1271 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: English | 1335 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: non-English | 1295 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: Chinese | 1347 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: German | 1307 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: Japanese | 1254 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: Korean | 1261 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: French | 1352 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: Spanish | 1317 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: Russian | 1291 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: Polish | 1299 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: multi-turn | 1307 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: longer queries | 1312 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: writing and language | 1290 | Not measured | Only Qwen3 30B A3B measured |
| LMArena Text: entertainment and media | 1258 | Not measured | Only Qwen3 30B A3B 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
| Benchmark | Qwen3 30B A3B | Reka Flash 3 | 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
Reka Flash 3 costs less
Repository review
50K input + 3K output tokens
Reka Flash 3 costs less
Agent loop
200K input + 10K output tokens
Reka Flash 3 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 | Qwen3 30B A3B | Reka Flash 3 |
|---|---|---|
| Maker | Qwen | Reka |
| Weights | Open weights | Open weights |
| Released | April 28, 2025 | March 12, 2025 |
| Context window | 41K | 66K |
| Input $/Mtok | $0.12 | $0.10 |
| Output $/Mtok | $0.50 | $0.20 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 30B A3B better than Reka Flash 3 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Compare them on price, context and the boards below.
Which is cheaper, Qwen3 30B A3B or Reka Flash 3?
Qwen3 30B A3B costs $0.12 per million input tokens and $0.50 per million output tokens; Reka Flash 3 costs $0.10 and $0.20. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3 30B A3B or Reka Flash 3?
Reka Flash 3, with 66K tokens against 41K.
How many benchmarks is the Qwen3 30B A3B vs Reka Flash 3 comparison based on?
11 benchmarks have a result for both models as of October 23, 2026. Qwen3 30B A3B's coding index rests on 0 independent benchmarks and Reka Flash 3's on 0. Where only one model has a result, the page says so instead of naming a winner.