Compare / DeepSeek-V4-Flash (0423) vs Hy3

DeepSeek-V4-Flash (0423) vs Hy3

Which is cheaper, which is more capable, and which does better at each kind of work, from 16 results on 2 sources that measured both models.

DeepSeekDeepSeek-V4-Flash (0423) Released 24 Apr 2026 · $0.050 in · $0.45 out
TencentHy3 Released 6 Jul 2026 · $0.14 in · $0.58 out
DeepSeek-V4-Flash (0423) wins 36 too close to callHy3 wins 7

Of the 16 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).

DeepSeek-V4-Flash (0423) or Hy3? The short answer

They score about the same, so pick DeepSeek-V4-Flash (0423): it costs 40% less per token.

  • DeepSeek-V4-Flash (0423) is cheaper: $0.15 against $0.25 per million tokens, 40% less.
  • They score about the same on the Artificial Analysis Intelligence Index (24.4 and 25.3).
  • Hy3 is newer: released 6 Jul 2026, against 24 Apr 2026 for DeepSeek-V4-Flash (0423).
  • On third-party benchmarks, on the results both models have: DeepSeek-V4-Flash (0423) does better for agents and tool use; Hy3 for reasoning and knowledge and writing.

Head to head

24.4
Intelligence Index
25.3
$0.15
Price per million tokens, blendedlower is better
$0.25
1,048,576
Context window, tokens
262,144

Every fact side by side

DeepSeek-V4-Flash (0423)Hy3
Price per million tokens$0.050 in · $0.45 out$0.14 in · $0.58 out
Blended (3 in : 1 out)$0.15$0.25
Cheapest host, blended$0.10 (StreamLake, FP8)$0.23 (Tencent, FP8)
Intelligence Index24.425.3
Spring Prompt overall––
Output speed, tokens a second–83
Context window1,048,576 tokens262,144 tokens
Released24 Apr 20266 Jul 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 59 of 159 70 of 159
Reasoning and knowledgeHard questions across science, maths and general knowledge 69 of 187 67 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 63 of 169 46 of 169
Sticking to the factsSummarising without inventing things, and factual answers 81 of 139 69 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.

MeasureDeepSeek-V4-Flash (0423)Hy3
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 85 · 58 of 387 measured 1,438high reasoning 1,456
Overall · Text factualityArena rating, higher is better 81 · 67 of 163 measured 1,439 1,445
Business, management and finance · TextArena rating, higher is better 91 · 69 of 380 measured 1,439 1,448
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 85 · 78 of 249 measured 24.4high reasoning 25.3
GPQA Diamond% of questions, higher is better 52 · 46 of 232 measured 89.4%max reasoning 89.7%
Humanity's Last Exam% of questions, higher is better 67 · 73 of 250 measured 34.8%max reasoning 33.5%
Show all 16 results (10 more)
MeasureDeepSeek-V4-Flash (0423)Hy3
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results
Creative writing · TextArena rating, higher is better 78 · 69 of 385 measured 1,410high reasoning 1,422
Expert prompts · TextArena rating, higher is better 83 · 55 of 338 measured 1,466high reasoning 1,488
Instruction following · TextArena rating, higher is better 79 · 61 of 387 measured 1,433high reasoning 1,446
Writing, literature and language · TextArena rating, higher is better 78 · 71 of 386 measured 1,422high reasoning 1,427
Artificial Analysis reported by Artificial Analysis · 6 results
Terminal-Bench 2.1% of tasks, higher is better 46 · 43 of 131 measured 61.8%max reasoning 64.4%
Terminal-Bench 4.0% of tasks, higher is better 52 · 65 of 120 measured 3.0%high reasoning 0.5%
τ-bench banking% of tasks, higher is better 34 · 44 of 121 measured 30.9%max reasoning 22.9%
Artificial Analysis Coding Indexindex score, higher is better 45 · 41 of 130 measured 56.2max reasoning 58.8
Long-context reasoning (AA-LCR)% of questions, higher is better 89 · 55 of 241 measured 74.3%max reasoning 79.0%
SciCode% of problems, higher is better 73 · 59 of 121 measured 45.3%max reasoning 48.6%

Quick answers

Is DeepSeek-V4-Flash (0423) better than Hy3?

They score about the same, so pick DeepSeek-V4-Flash (0423): it costs 40% less per token.

Which is cheaper, DeepSeek-V4-Flash (0423) or Hy3?

DeepSeek-V4-Flash (0423) is cheaper: $0.15 against $0.25 per million tokens, 40% less.

Which is better for agents and tool use, DeepSeek-V4-Flash (0423) or Hy3?

DeepSeek-V4-Flash (0423) ranks 59 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 70 for Hy3.

Which is better for reasoning and knowledge, DeepSeek-V4-Flash (0423) or Hy3?

Hy3 ranks 67 of 187 models on sale for hard questions across science, maths and general knowledge, against 69 for DeepSeek-V4-Flash (0423).

Which is better for writing, DeepSeek-V4-Flash (0423) or Hy3?

Hy3 ranks 46 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 63 for DeepSeek-V4-Flash (0423).

Which is better for sticking to the facts, DeepSeek-V4-Flash (0423) or Hy3?

Hy3 ranks 69 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, DeepSeek-V4-Flash (0423) or Hy3?

DeepSeek-V4-Flash (0423): 1,048,576 tokens, against 262,144 for Hy3.

Run DeepSeek-V4-Flash (0423) and Hy3 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.