Qwen3 14B vs Qwen3.7 Flash
Qwen3.7 Flash costs 1.8x less per output token than Qwen3 14B ($0.13 vs $0.23 per million). Updated October 21, 2026; 3 benchmarks have a result for both.
Qwen3 14B or Qwen3.7 Flash: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen3 14B for work measured by Epoch Capabilities Index or GPQA Diamond (Epoch AI run), where it leads.
- Shared results
- 3
- Qwen3 14B only
- 17
- Qwen3.7 Flash only
- 3
- Like-for-like categories
- 3 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 Qwen3 14B has a AA Intelligence Index result, 6.7.
Directional only
Long documents
Listed context window
Qwen3.7 Flash
Qwen3.7 Flash takes 1M tokens against 41K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3.7 Flash
Qwen3 14B $0.0002, Qwen3.7 Flash $0.0001 per turn.
List prices
At a glance
- Coding index
- Qwen3 14B not benchmarked, Qwen3.7 Flash not benchmarked.
- Widest category
- Math, 20.3 points apart on average over 1 shared benchmark.
- Widest single gap
- OTIS Mock AIME: Qwen3 14B 66.4% against Qwen3.7 Flash 86.7%, 20.3 points apart on a 0 to 100 scale.
- Output price
- Qwen3 14B $0.23 against Qwen3.7 Flash $0.13 per million output tokens, 1.8x apart.
- Context window
- Qwen3 14B takes 41K tokens, Qwen3.7 Flash 1M.
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 14B | Bars | Qwen3.7 Flash | Ties |
|---|---|---|---|---|
| Intelligence | 0 | 1 | ||
| Reasoning | 0 | 1 | ||
| Math | 0 | 1 | ||
| All shared boards | 0 | 3 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index4 boards, none shared
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| LiveCodeBench (Artificial Analysis run) | 52.3% | Not measured | Only Qwen3 14B measured |
| SciCode | 30.7% | Not measured | Only Qwen3 14B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 5.3% | Not measured | Only Qwen3 14B measured |
| ALE-Bench | Not measured | 420.1 | Only Qwen3.7 Flash measured |
| Neither model has a result on the other 35 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| AA Intelligence Index | 6.7 | Not measured | Only Qwen3 14B measured |
| Epoch Capabilities Index | 138.2 | 144.1 | Qwen3.7 Flash |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 34.5% | Not measured | Only Qwen3 14B measured |
| Berkeley Function Calling Leaderboard (BFCL) | 41.0% | Not measured | Only Qwen3 14B measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 60.4% | Not measured | Only Qwen3 14B measured |
| Humanity's Last Exam (Artificial Analysis run) | 4.5% | Not measured | Only Qwen3 14B measured |
| GPQA Diamond (Epoch AI run) | 63.8% | 82.3% | Qwen3.7 Flash |
| Mystery game puzzles | Not measured | 15.0% | Only Qwen3.7 Flash measured |
| Kagi LLM Benchmark | 49.1% | Not measured | Only Qwen3 14B measured |
| Neither model has a result on the other 17 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| AA Math Index | 58.0 | Not measured | Only Qwen3 14B measured |
| AIME 2025 (Artificial Analysis run) | 58.0% | Not measured | Only Qwen3 14B measured |
| OTIS Mock AIME | 66.4% | 86.7% | Qwen3.7 Flash |
| FrontierMath, tiers 1 to 3 | Not measured | 19.3% | Only Qwen3.7 Flash measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 77.4% | Not measured | Only Qwen3 14B measured |
| Neither model has a result on the other 17 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Qwen3 14B | Qwen3.7 Flash | Leader |
|---|---|---|---|
| IFBench | 40.5% | Not measured | Only Qwen3 14B measured |
| Long context reasoning (AA-LCR) | 0.0% | Not measured | Only Qwen3 14B measured |
| DTBench | 64.0% | Not measured | Only Qwen3 14B measured |
| LMCA | 18.2% | Not measured | Only Qwen3 14B measured |
| Fiction.LiveBench | 62.5% | Not measured | Only Qwen3 14B measured |
| Neither model has a result on the other 41 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3 14B | Qwen3.7 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
Qwen3.7 Flash costs less
Repository review
50K input + 3K output tokens
Qwen3.7 Flash costs less
Agent loop
200K input + 10K output tokens
Qwen3.7 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 | Qwen3 14B | Qwen3.7 Flash |
|---|---|---|
| Maker | Qwen | Qwen |
| Weights | Open weights | No public weights listed |
| Released | April 28, 2025 | July 27, 2026 |
| Context window | 41K | 1M |
| Input $/Mtok | $0.11 | $0.03 |
| Output $/Mtok | $0.23 | $0.13 |
| 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 14B better than Qwen3.7 Flash for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen3 14B for work measured by Epoch Capabilities Index or GPQA Diamond (Epoch AI run), where it leads.
Which is cheaper, Qwen3 14B or Qwen3.7 Flash?
Qwen3 14B costs $0.11 per million input tokens and $0.23 per million output tokens; Qwen3.7 Flash costs $0.03 and $0.13. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3 14B or Qwen3.7 Flash?
Qwen3.7 Flash, with 1M tokens against 41K.
How many benchmarks is the Qwen3 14B vs Qwen3.7 Flash comparison based on?
3 benchmarks have a result for both models as of October 21, 2026. Qwen3 14B's coding index rests on 0 independent benchmarks and Qwen3.7 Flash's on 0. Where only one model has a result, the page says so instead of naming a winner.