DeepSeek V4 Flash 0423 vs Qwen3 VL 8B Instruct
DeepSeek V4 Flash 0423 leads Qwen3 VL 8B Instruct on the Artificial Analysis Intelligence Index, 24.4 to 7.3, and DeepSeek V4 Flash 0423 costs 2.7x less per output token ($0.28 vs $0.75 per million). Updated October 21, 2026; 1 benchmarks have a result for both.
DeepSeek V4 Flash 0423 or Qwen3 VL 8B Instruct: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick DeepSeek V4 Flash 0423 for output tokens 2.7x cheaper ($0.28 vs $0.75 per million) and a larger context window (1.05M against 262K tokens).
- Shared results
- 1
- DeepSeek V4 Flash 0423 only
- 48
- Qwen3 VL 8B Instruct only
- 10
- Like-for-like categories
- 1 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
DeepSeek V4 Flash 0423
DeepSeek V4 Flash 0423 leads, 24.4 against 7.3.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Flash 0423 has a AA Agentic Index result, 26.3.
Directional only
Web design
Design Arena: websites
No call
Only DeepSeek V4 Flash 0423 has a Design Arena: websites result, 1208.
Directional only
Long documents
Listed context window
DeepSeek V4 Flash 0423
DeepSeek V4 Flash 0423 takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0423
DeepSeek V4 Flash 0423 $0.0003, Qwen3 VL 8B Instruct $0.0006 per turn.
List prices
At a glance
- Coding index
- DeepSeek V4 Flash 0423 not benchmarked, Qwen3 VL 8B Instruct not benchmarked.
- Widest category
- Intelligence, 17.1 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Intelligence Index: DeepSeek V4 Flash 0423 24.4 against Qwen3 VL 8B Instruct 7.3, 17.1 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0423 $0.28 against Qwen3 VL 8B Instruct $0.75 per million output tokens, 2.7x apart.
- Context window
- DeepSeek V4 Flash 0423 takes 1.05M tokens, Qwen3 VL 8B Instruct 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 | DeepSeek V4 Flash 0423 | Bars | Qwen3 VL 8B Instruct | Ties |
|---|---|---|---|---|
| Intelligence | 1 | 0 | ||
| All shared boards | 1 | 0 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index10 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| LMArena WebDev | 1582 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| AA Coding Index | 52.0 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 35.3% | Only Qwen3 VL 8B Instruct measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 3.8% | Only Qwen3 VL 8B Instruct measured |
| WeirdML | 45.6% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| ALE-Bench | 678.2 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: coding | 1457 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: software and IT | 1453 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena WebDev: React | 1585 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena WebDev: HTML | 1558 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 29 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.4 | 7.3 | DeepSeek V4 Flash 0423 |
| Epoch Capabilities Index | 146.1 | Not measured | Only DeepSeek V4 Flash 0423 measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| AA Agentic Index | 26.3 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| τ²-Bench Telecom | Not measured | 29.2% | Only Qwen3 VL 8B Instruct measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 57.9% | Only Qwen3 VL 8B Instruct measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 3.8% | Only Qwen3 VL 8B Instruct measured |
| SimpleBench | 46.3% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: hard prompts | 1444 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: expert | 1441 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: hard prompts in English | 1449 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Kagi LLM Benchmark | 52.2% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 15 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| AA Math Index | Not measured | 30.7 | Only Qwen3 VL 8B Instruct measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 30.7% | Only Qwen3 VL 8B Instruct measured |
| LMArena Text: math | 1426 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: mathematical work | 1432 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | Not measured | 74.9% | Only Qwen3 VL 8B Instruct measured |
| LMArena Text: science | 1446 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: medicine and healthcare | 1448 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: legal and government | 1439 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: business and finance | 1425 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 13 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| Design Arena: websites | 1208 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: UI components | 1165 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: game dev | 1197 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: data viz | 1127 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: 3D | 1189 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: all code categories | 1205 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: SVG | 1155 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: ASCII art | 1109 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 1 board here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index21 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | Leader |
|---|---|---|---|
| IFBench | Not measured | 39.9% | Only Qwen3 VL 8B Instruct measured |
| Long context reasoning (AA-LCR) | Not measured | 33.3% | Only Qwen3 VL 8B Instruct measured |
| DTBench | 86.4% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMCA | 35.9% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text | 1432 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: instruction following | 1421 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: creative writing | 1403 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: English | 1441 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: non-English | 1419 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Chinese | 1463 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: German | 1419 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Japanese | 1389 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Korean | 1383 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: French | 1439 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Spanish | 1436 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Russian | 1427 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Polish | 1436 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: multi-turn | 1449 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: longer queries | 1434 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: writing and language | 1410 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: entertainment and media | 1400 | Not measured | Only DeepSeek V4 Flash 0423 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 | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct | 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
DeepSeek V4 Flash 0423 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Flash 0423 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Flash 0423 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 | DeepSeek V4 Flash 0423 | Qwen3 VL 8B Instruct |
|---|---|---|
| Maker | DeepSeek | Qwen |
| Weights | Open weights | Open weights |
| Released | April 24, 2026 | October 14, 2025 |
| Context window | 1.05M | 262K |
| Input $/Mtok | $0.13 | $0.25 |
| Output $/Mtok | $0.28 | $0.75 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 1 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0423 better than Qwen3 VL 8B Instruct for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick DeepSeek V4 Flash 0423 for output tokens 2.7x cheaper ($0.28 vs $0.75 per million) and a larger context window (1.05M against 262K tokens).
Which is cheaper, DeepSeek V4 Flash 0423 or Qwen3 VL 8B Instruct?
DeepSeek V4 Flash 0423 costs $0.13 per million input tokens and $0.28 per million output tokens; Qwen3 VL 8B Instruct costs $0.25 and $0.75. Cost per solved task is known for neither of them.
Which has the bigger context window, DeepSeek V4 Flash 0423 or Qwen3 VL 8B Instruct?
DeepSeek V4 Flash 0423, with 1.05M tokens against 262K.
How many benchmarks is the DeepSeek V4 Flash 0423 vs Qwen3 VL 8B Instruct comparison based on?
1 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Flash 0423's coding index rests on 1 independent benchmark and Qwen3 VL 8B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.