Compare / Qwen3.5-397B-A17B vs DeepSeek-V4-Flash (0423)

Qwen3.5-397B-A17B vs DeepSeek-V4-Flash (0423)

Which is cheaper, which is more capable, and which does better at each kind of work, from 22 results on 3 sources that measured both models.

AlibabaQwen3.5-397B-A17B Released 13 Feb 2026 · $0.55 in · $3.50 out
DeepSeekDeepSeek-V4-Flash (0423) Released 24 Apr 2026 · $0.050 in · $0.45 out
Qwen3.5-397B-A17B wins 28 too close to callDeepSeek-V4-Flash (0423) wins 12

Of the 22 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also Alibaba vs DeepSeek, every model from both labs.

Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)? The short answer

Pick DeepSeek-V4-Flash (0423): it scores higher (Intelligence Index 24.4 against 21.4) and costs less.

  • DeepSeek-V4-Flash (0423) is cheaper: $0.15 against $1.29 per million tokens, 8.6 times less.
  • DeepSeek-V4-Flash (0423) scores higher on the Artificial Analysis Intelligence Index: 24.4 against 21.4.
  • DeepSeek-V4-Flash (0423) is newer: released 24 Apr 2026, against 13 Feb 2026 for Qwen3.5-397B-A17B.
  • On third-party benchmarks, on the results both models have: DeepSeek-V4-Flash (0423) does better for agents and tool use, reasoning and knowledge and writing.

Head to head

21.4
Intelligence Index
24.4
$1.29
Price per million tokens, blendedlower is better
$0.15
262,144
Context window, tokens
1,048,576

Every fact side by side

Qwen3.5-397B-A17BDeepSeek-V4-Flash (0423)
Price per million tokens$0.55 in · $3.50 out$0.050 in · $0.45 out
Blended (3 in : 1 out)$1.29$0.15
Cheapest host, blended$0.88 (Alibaba)$0.10 (StreamLake, FP8)
Intelligence Index21.424.4
Spring Prompt overall––
Output speed, tokens a second56–
Context window262,144 tokens1,048,576 tokens
Released13 Feb 202624 Apr 2026 (newer)
WeightsOpenOpen
Developer based inChinaChina

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 71 of 159 59 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 48 of 51 not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge 82 of 187 69 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 68 of 169 63 of 169
Sticking to the factsSummarising without inventing things, and factual answers 74 of 139 81 of 139

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureQwen3.5-397B-A17BDeepSeek-V4-Flash (0423)
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 80 · 85 of 387 measured 1,442 1,438high reasoning
Overall · Text factualityArena rating, higher is better 70 · 81 of 163 measured 1,445 1,439
Business, management and finance · TextArena rating, higher is better 73 · 91 of 380 measured 1,446 1,439
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 103 · 85 of 249 measured 21.4no reasoning 24.4high reasoning
GPQA Diamond% of questions, higher is better 53 · 52 of 232 measured 89.3% 89.4%max reasoning
Humanity's Last Exam% of questions, higher is better 85 · 67 of 250 measured 29.0% 34.8%max reasoning
UGI Leaderboard reported by UGI Leaderboard · 3 results
Writing score · UGIscore out of 100, higher is better 77 · 60 of 230 measured 50.1no reasoning 54.6no reasoning
Requested-length error · UGI% off the requested word count, lower is better 134 · 54 of 230 measured 23.0%no reasoning 12.0%thinking reasoning
Style adherence · UGIscore from 0 to 1, higher is better 164 · 55 of 230 measured 0.33no reasoning 0.37thinking reasoning
Show all 22 results (13 more)
MeasureQwen3.5-397B-A17BDeepSeek-V4-Flash (0423)
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results
Creative writing · TextArena rating, higher is better 86 · 78 of 385 measured 1,405 1,410high reasoning
Expert prompts · TextArena rating, higher is better 67 · 83 of 338 measured 1,477 1,466high reasoning
Instruction following · TextArena rating, higher is better 78 · 79 of 387 measured 1,434 1,433high reasoning
Writing, literature and language · TextArena rating, higher is better 80 · 78 of 386 measured 1,421 1,422high reasoning
Artificial Analysis reported by Artificial Analysis · 9 results
Terminal-Bench 2.1% of tasks, higher is better 62 · 46 of 131 measured 51.3% 61.8%max reasoning
Terminal-Bench 4.0% of tasks, higher is better 72 · 52 of 120 measured 0.0% 3.0%high reasoning
τ-bench banking% of tasks, higher is better 64 · 34 of 121 measured 13.4% 30.9%max reasoning
Artificial Analysis Coding Indexindex score, higher is better 60 · 45 of 130 measured 48.2 56.2max reasoning
IFBench% of instructions, higher is better 8 · 7 of 195 measured 78.8% 79.2%max reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 73 · 89 of 241 measured 77.3% 74.3%max reasoning
SciCode% of problems, higher is better 75 · 73 of 121 measured 44.8% 45.3%max reasoning
Terminal-Bench Hard% of tasks, higher is better 34 · 37 of 190 measured 40.9% 38.6%high reasoning
τ²-bench telecom% of tasks, higher is better 16 · 16 of 193 measured 95.6% 95.6%high reasoning

Quick answers

Is Qwen3.5-397B-A17B better than DeepSeek-V4-Flash (0423)?

Pick DeepSeek-V4-Flash (0423): it scores higher (Intelligence Index 24.4 against 21.4) and costs less.

Which is cheaper, Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)?

DeepSeek-V4-Flash (0423) is cheaper: $0.15 against $1.29 per million tokens, 8.6 times less.

Which is better for agents and tool use, Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)?

DeepSeek-V4-Flash (0423) ranks 59 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 71 for Qwen3.5-397B-A17B.

Which is better for reasoning and knowledge, Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)?

DeepSeek-V4-Flash (0423) ranks 69 of 187 models on sale for hard questions across science, maths and general knowledge, against 82 for Qwen3.5-397B-A17B.

Which is better for writing, Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)?

DeepSeek-V4-Flash (0423) ranks 63 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 68 for Qwen3.5-397B-A17B.

Which is better for sticking to the facts, Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)?

Qwen3.5-397B-A17B ranks 74 of 139 models on sale for summarising without inventing things, and factual answers, against 81 for DeepSeek-V4-Flash (0423).

Which has the bigger context window, Qwen3.5-397B-A17B or DeepSeek-V4-Flash (0423)?

DeepSeek-V4-Flash (0423): 1,048,576 tokens, against 262,144 for Qwen3.5-397B-A17B.

Run Qwen3.5-397B-A17B and DeepSeek-V4-Flash (0423) on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.