Enter opens the highlighted result. Esc closes.

Ministral 3 14B 2512 vs Qwen2.5 VL 72B Instruct

Ministral 3 14B 2512 costs 5.0x less per output token than Qwen2.5 VL 72B Instruct ($0.20 vs $1.00 per million). Updated October 23, 2026; no benchmark has tested both models yet.

Replace Ministral 3 14B 2512 Replace Qwen2.5 VL 72B Instruct
Ministral 3 14B 2512Mistral AI
Coding
No data
Overall
No data
Output $/M
$0.20
Qwen2.5 VL 72B InstructQwen
Coding
No data
Overall
No data
Output $/M
$1.00
vs $0.20 vs $1.00 per 1M output tokens. Ministral 3 14B 2512 is 5.0x cheaper.

Ministral 3 14B 2512 or Qwen2.5 VL 72B Instruct: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick Ministral 3 14B 2512 for output tokens 5.0x cheaper ($0.20 vs $1.00 per million) and a larger context window (262K against 128K tokens).

Model AMinistral 3 14B 2512Mistral AINot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen2.5 VL 72B InstructQwenNot benchmarkedNo coding benchmarkNo cost per solved task

No benchmark has tested both yet. Ministral 3 14B 2512 has results on AA Coding Index, AA Intelligence Index and AA Agentic Index; Qwen2.5 VL 72B Instruct on OSWorld, Kagi LLM Benchmark and GeoBench. Compare them on price and context instead.

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.

At a glance

Coding index
Ministral 3 14B 2512 not benchmarked, Qwen2.5 VL 72B Instruct not benchmarked.
Widest category
None yet: no category has been tested on both models.
Widest single gap
None yet: no benchmark has tested both models.
Output price
Ministral 3 14B 2512 $0.20 against Qwen2.5 VL 72B Instruct $1.00 per million output tokens, 5.0x apart.
Context window
Ministral 3 14B 2512 takes 262K tokens, Qwen2.5 VL 72B Instruct 128K.

Benchmark by benchmark

Every board either model has a result on, a category to a fold; bars are each result on a 0 to 100 scale. Only one model is measured on each, so no row names a leader.

CodingDirectional onlySix of these boards feed the Vibe Coding Index1 board measured
Coding benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
AA Coding Index14.4Not measuredOnly Ministral 3 14B 2512 measured
Neither model has a result on the other 42 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board measured
Intelligence benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
AA Intelligence Index6.0Not measuredOnly Ministral 3 14B 2512 measured
Neither model has a result on the other 1 board here
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards measured
Agentic work benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
AA Agentic Index1.1Not measuredOnly Ministral 3 14B 2512 measured
OSWorldNot measured5.0%Only Qwen2.5 VL 72B Instruct measured
Neither model has a result on the other 40 boards here
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index1 board measured
Reasoning benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
Kagi LLM BenchmarkNot measured36.0%Only Qwen2.5 VL 72B Instruct measured
Neither model has a result on the other 22 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index1 board measured
Knowledge benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
GeoBenchNot measured62.0%Only Qwen2.5 VL 72B Instruct measured
Neither model has a result on the other 19 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index4 boards measured
Web design benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
Design Arena: websites1081Not measuredOnly Ministral 3 14B 2512 measured
Design Arena: game dev1037Not measuredOnly Ministral 3 14B 2512 measured
Design Arena: 3D991Not measuredOnly Ministral 3 14B 2512 measured
Design Arena: all code categories1064Not measuredOnly Ministral 3 14B 2512 measured
Neither model has a result on the other 5 boards here
GeneralDirectional onlyPublic boards, shown as published, kept out of the index3 boards measured
General benchmarks for Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
BenchmarkMinistral 3 14B 2512Qwen2.5 VL 72B InstructLeader
LMArena VisionNot measured1107Only Qwen2.5 VL 72B Instruct measured
LMArena Vision: EnglishNot measured1115Only Qwen2.5 VL 72B Instruct measured
LMArena Vision: ChineseNot measured1107Only Qwen2.5 VL 72B Instruct measured
Neither model has a result on the other 45 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

Ministral 3 14B 2512$0.0003Fits in one request
Qwen2.5 VL 72B Instruct$0.0013Fits in one request

Ministral 3 14B 2512 costs less

Repository review

50K input + 3K output tokens

Ministral 3 14B 2512$0.011Fits in one request
Qwen2.5 VL 72B Instruct$0.043Fits in one request

Ministral 3 14B 2512 costs less

Agent loop

200K input + 10K output tokens

Ministral 3 14B 2512$0.042Fits in one request
Qwen2.5 VL 72B Instruct$0.170Exceeds the context window

Ministral 3 14B 2512 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.

Specifications of Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct
SpecificationMistral Ministral 3 14B 2512Qwen2.5 VL 72B Instruct
MakerMistral AIQwen
WeightsOpen weightsOpen weights
ReleasedDecember 2, 2025February 1, 2025
Context window262K128K
Input $/Mtok$0.20$0.80
Output $/Mtok$0.20$1.00
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Ministral 3 14B 2512 better than Qwen2.5 VL 72B Instruct for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Ministral 3 14B 2512 for output tokens 5.0x cheaper ($0.20 vs $1.00 per million) and a larger context window (262K against 128K tokens).

Which is cheaper, Ministral 3 14B 2512 or Qwen2.5 VL 72B Instruct?

Ministral 3 14B 2512 costs $0.20 per million input tokens and $0.20 per million output tokens; Qwen2.5 VL 72B Instruct costs $0.80 and $1.00. Cost per solved task is known for neither of them.

Which has the bigger context window, Ministral 3 14B 2512 or Qwen2.5 VL 72B Instruct?

Ministral 3 14B 2512, with 262K tokens against 128K.

Has any benchmark tested both Ministral 3 14B 2512 and Qwen2.5 VL 72B Instruct?

No benchmark has tested both yet. Ministral 3 14B 2512 has results on AA Coding Index, AA Intelligence Index and AA Agentic Index; Qwen2.5 VL 72B Instruct on OSWorld, Kagi LLM Benchmark and GeoBench. Compare them on price and context instead. Data as of October 23, 2026.

Compare next