Noometry

Model comparison

Qwen3-VL 235B-A22B vs Step 3.7 Flash

Qwen3-VL 235B-A22B is the stronger model overall, scoring 43.2 to 37.3 on the Noometry Index. Step 3.7 Flash costs 2.9× less per token, which makes it the better buy when Qwen3-VL 235B-A22B's lead doesn't matter for your workload.

Last verified . 0 shared benchmarks.

Qwen3-VL 235B-A22B Alibaba (Qwen)

43.2

Rank #95 Confirmed

Step 3.7 Flash StepFun

37.3

Rank #207 Reported

Summary

  • The widest gap is in reasoning, where Qwen3-VL 235B-A22B leads 29.3 to 21.6.
  • Step 3.7 Flash is cheaper at $0.18 / $1.11 per million input/output tokens, against $0.70 / $2.80 for Qwen3-VL 235B-A22B.
  • Step 3.7 Flash accepts more context: 256K tokens versus 131K.

Side by side

Qwen3-VL 235B-A22BStep 3.7 Flash
ProviderAlibaba (Qwen)StepFun
Noometry Index43.237.3
Released2025-042026-05-29
WeightsOpenOpen
Context window131K256K
Max output33K256K
Input $ / M tokens$0.70$0.18
Output $ / M tokens$2.80$1.11
Results tracked185

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Qwen3-VL 235B-A22B leads

Qwen3-VL 235B-A22B: 42.4 (#100), Step 3.7 Flash: 40.0 (#150)

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
SciCode—40%
LMArena Coding1439—
ALE-Bench—694.12

Reasoning Qwen3-VL 235B-A22B leads

Qwen3-VL 235B-A22B: 29.3 (#92), Step 3.7 Flash: 21.6 (#219)

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
NYT Connections (extended)—39.7%
CritPt—2.3%
LMArena Hard Prompts1428—

Math Step 3.7 Flash leads

Qwen3-VL 235B-A22B: 39.0 (#118), Step 3.7 Flash: 42.9 (#82)

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
MathArena Final-Answer Competitions—68.5%
LMArena Math1426—

Knowledge Not comparable

Qwen3-VL 235B-A22B: 40.3 (#121), Step 3.7 Flash: —

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
LMArena Expert1442—

Multimodal Not comparable

Qwen3-VL 235B-A22B: 39.8 (#55), Step 3.7 Flash: —

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
LMArena Vision1247—

Multilingual Not comparable

Qwen3-VL 235B-A22B: 51.9 (#97), Step 3.7 Flash: —

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
LMArena Non-English1405—
LMArena Chinese1463—
LMArena French1452—
LMArena German1424—
LMArena Japanese1385—
LMArena Korean1394—
LMArena Russian1408—
LMArena Spanish1428—

Instruction Following Not comparable

Qwen3-VL 235B-A22B: 74.2 (#101), Step 3.7 Flash: —

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
LMArena Instruction Following1406—

Long Context Not comparable

Qwen3-VL 235B-A22B: 43.4 (#98), Step 3.7 Flash: —

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
LMArena Longer Query1420—

Writing & Preference Not comparable

Qwen3-VL 235B-A22B: 60.2 (#99), Step 3.7 Flash: —

BenchmarkQwen3-VL 235B-A22BStep 3.7 Flash
LMArena Text1420—
LMArena Creative Writing1366—
LMArena Multi-Turn1428—

Frequently asked questions

Is Qwen3-VL 235B-A22B better than Step 3.7 Flash?

Qwen3-VL 235B-A22B is the stronger model overall, scoring 43.2 to 37.3 on the Noometry Index. Step 3.7 Flash costs 2.9× less per token, which makes it the better buy when Qwen3-VL 235B-A22B's lead doesn't matter for your workload.

Which is cheaper, Qwen3-VL 235B-A22B or Step 3.7 Flash?

Step 3.7 Flash is cheaper. It lists at $0.18 per million input tokens and $1.11 per million output tokens; Qwen3-VL 235B-A22B lists at $0.70 and $2.80.

Is Qwen3-VL 235B-A22B or Step 3.7 Flash better for coding?

Qwen3-VL 235B-A22B scores higher on coding benchmarks: 42.4 versus 40.0 in the Noometry coding category.

Which has the bigger context window?

Step 3.7 Flash does, with 256K tokens against 131K.

How many benchmarks do Qwen3-VL 235B-A22B and Step 3.7 Flash share?

0 benchmarks have published results for both models. Qwen3-VL 235B-A22B has 18 scored results on Noometry and Step 3.7 Flash has 5.

Related comparisons

Go deeper