Compare · ModelsLive · 3 picked · head to head
DeepSeek V3.2 Speciale vs MiMo-V2-Flash vs Qwen3.5 397B A17B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Qwen3.5 397B A17B wins on 6/11 benchmarks
Qwen3.5 397B A17B wins 6 of 11 shared benchmarks. Leads in knowledge · coding · arena.
Category leads
speed·MiMo-V2-Flashmath·DeepSeek V3.2 Specialeknowledge·Qwen3.5 397B A17Blanguage·DeepSeek V3.2 Specialecoding·Qwen3.5 397B A17Barena·Qwen3.5 397B A17B
Hype vs Reality
Attention vs performance
DeepSeek V3.2 Speciale
#6 by perf·no signal
MiMo-V2-Flash
#13 by perf·no signal
Qwen3.5 397B A17B
#48 by perf·#2 by attention
Best value
MiMo-V2-Flash
3.9x better value than DeepSeek V3.2 Speciale
DeepSeek V3.2 Speciale
97.8 pts/$
$0.80/M
MiMo-V2-Flash
385.8 pts/$
$0.19/M
Qwen3.5 397B A17B
29.6 pts/$
$2.02/M
Vendor risk
Mixed exposure
One or more vendors flagged
DeepSeek
$3.4B·Tier 1
Xiaomi (MiMo)
$111.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
11 benchmarks · 3 models
DeepSeek V3.2 SpecialeMiMo-V2-FlashQwen3.5 397B A17B
Artificial Analysis · Agentic Index
MiMo-V2-Flash leads by +28.9
Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?"
DeepSeek V3.2 Speciale
0.0
MiMo-V2-Flash
48.8
Qwen3.5 397B A17B
19.9
Artificial Analysis · Coding Index
Qwen3.5 397B A17B leads by +10.3
Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads.
DeepSeek V3.2 Speciale
37.9
MiMo-V2-Flash
33.5
Qwen3.5 397B A17B
48.2
Artificial Analysis · Quality Index
MiMo-V2-Flash leads by +12.0
DeepSeek V3.2 Speciale
29.4
MiMo-V2-Flash
41.5
Qwen3.5 397B A17B
21.4
OpenCompass · AIME2025
DeepSeek V3.2 Speciale leads by +3.1
DeepSeek V3.2 Speciale
96.0
MiMo-V2-Flash
92.9
Qwen3.5 397B A17B
92.3
OpenCompass · GPQA-Diamond
Qwen3.5 397B A17B leads by +1.7
DeepSeek V3.2 Speciale
86.7
MiMo-V2-Flash
82.1
Qwen3.5 397B A17B
88.4
OpenCompass · HLE
DeepSeek V3.2 Speciale leads by +1.1
DeepSeek V3.2 Speciale
28.6
MiMo-V2-Flash
20.5
Qwen3.5 397B A17B
27.5
OpenCompass · IFEval
DeepSeek V3.2 Speciale leads by +0.2
DeepSeek V3.2 Speciale
91.7
MiMo-V2-Flash
89.5
Qwen3.5 397B A17B
91.5
OpenCompass · LiveCodeBenchV6
Qwen3.5 397B A17B leads by +2.1
DeepSeek V3.2 Speciale
80.9
MiMo-V2-Flash
71.8
Qwen3.5 397B A17B
83.0
OpenCompass · MMLU-Pro
Qwen3.5 397B A17B leads by +2.1
DeepSeek V3.2 Speciale
85.5
MiMo-V2-Flash
83.1
Qwen3.5 397B A17B
87.6
Chatbot Arena Elo · Coding
Qwen3.5 397B A17B leads by +62.9
MiMo-V2-Flash
1336.5
Qwen3.5 397B A17B
1399.4
Chatbot Arena Elo · Overall
Qwen3.5 397B A17B leads by +49.7
MiMo-V2-Flash
1392.1
Qwen3.5 397B A17B
1441.8
Full benchmark table
| Benchmark | DeepSeek V3.2 Speciale | MiMo-V2-Flash | Qwen3.5 397B A17B |
|---|---|---|---|
Artificial Analysis · Agentic Index Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?" | 0.0 | 48.8 | 19.9 |
Artificial Analysis · Coding Index Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads. | 37.9 | 33.5 | 48.2 |
Artificial Analysis · Quality Index | 29.4 | 41.5 | 21.4 |
OpenCompass · AIME2025 | 96.0 | 92.9 | 92.3 |
OpenCompass · GPQA-Diamond | 86.7 | 82.1 | 88.4 |
OpenCompass · HLE | 28.6 | 20.5 | 27.5 |
OpenCompass · IFEval | 91.7 | 89.5 | 91.5 |
OpenCompass · LiveCodeBenchV6 | 80.9 | 71.8 | 83.0 |
OpenCompass · MMLU-Pro | 85.5 | 83.1 | 87.6 |
Chatbot Arena Elo · Coding | — | 1336.5 | 1399.4 |
Chatbot Arena Elo · Overall | — | 1392.1 | 1441.8 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.40 | $1.20 | 164K tokens (~82 books) | $6.00 | |
| $0.09 | $0.29 | 262K tokens (~131 books) | $1.40 | |
| $0.55 | $3.50 | 262K tokens (~131 books) | $12.88 |