Best LLMs for Instruction Following — Qwen

Match · Updated August 2026

The best Qwen model for instruction following is Qwen3.7 Max, satisfying 80.5% of IFBench's verifiable prompt constraints. Qwen3.5 397B A17B (78.8%) and Qwen3.7 Plus (78.0%) round out the top three.

80.5%IFBench

46.0Intelligence

207 t/sSpeed

$1.48Input /M

1MContext

1
qwen3.7-max
ReasoningToolsJSON46.0 intel · $1.48/M · 207 t/s
80.5%
IFBench
2
qwen3.5-397b-a17b
ReasoningToolsJSON+133.7 intel · $0.390/M · 69 t/s
78.8%
IFBench
3
qwen3.7-plus
ReasoningToolsJSON+139.0 intel · $0.320/M · 12 t/s
78.0%
IFBench
4
qwen3.6-max-preview
ReasoningToolsJSON40.0 intel · $1.03/M · 262K ctx
76.6%
IFBench
5
qwen3.5-122b-a10b
ReasoningToolsJSON+132.3 intel · $0.260/M · 127 t/s
75.7%
IFBench
6
qwen3.5-27b
ReasoningToolsJSON+133.8 intel · $0.195/M · 262K ctx
75.6%
IFBench
7
qwen3.6-plus
ReasoningToolsJSON+139.6 intel · $0.325/M · 55 t/s
75.2%
IFBench
8
qwen3.5-35b-a3b
ReasoningToolsJSON+129.3 intel · $0.140/M · 262K ctx
72.5%
IFBench
9
qwen3-max-thinking
ReasoningToolsJSON31.7 intel · $0.780/M · 262K ctx
70.7%
IFBench
10
qwen3.6-27b
ReasoningToolsJSON+137.1 intel · $0.289/M · 60 t/s
67.6%
IFBench
11
qwen3.5-9b
ReasoningToolsJSON+121.4 intel · $0.100/M · 74 t/s
66.7%
IFBench
12
qwen3.6-35b-a3b
ReasoningToolsJSON+131.6 intel · $0.140/M · 158 t/s
64.4%
IFBench
13
qwen3-max
ToolsJSON24.0 intel · $0.780/M · 29 t/s
44.1%
IFBench
14
qwen3-vl-235b-a22b-instruct
ToolsJSONVision14.3 intel · $0.210/M · 19 t/s
42.7%
IFBench
15
qwen3-next-80b-a3b-instruct
ToolsJSON13.7 intel · $0.090/M · 74 t/s
39.7%
IFBench
16
qwen3-vl-32b-instruct
ToolsJSONVision11.1 intel · $0.104/M · 38 t/s
39.2%
IFBench
17
qwen-2.5-72b-instruct
ToolsJSON9.6 intel · $0.360/M · 33K ctx
36.9%
IFBench
18
qwen3-coder-next
ToolsJSON21.1 intel · $0.120/M · 133 t/s
35.2%
IFBench
19
qwen3-vl-30b-a3b-instruct
ToolsJSONVision10.0 intel · $0.150/M · 34 t/s
33.1%
IFBench
20
qwen3-coder-30b-a3b-instruct
ToolsJSON13.6 intel · $0.070/M · 81 t/s
32.7%
IFBench
21
qwen3-vl-8b-instruct
ToolsJSONVision8.4 intel · $0.117/M · 42 t/s
32.3%
IFBench

How this is ranked

AI models ranked by IFBench — how reliably a model obeys explicit, verifiable constraints in the prompt (format, length, inclusion and exclusion rules). High scores mean fewer retries and less prompt-wrangling in production, which often matters more than raw intelligence for pipelines.

Frequently asked

Which Qwen model follows instructions most reliably?

The best Qwen model for instruction following is Qwen3.7 Max, satisfying 80.5% of IFBench's verifiable prompt constraints. Qwen3.5 397B A17B (78.8%) and Qwen3.7 Plus (78.0%) round out the top three.

What's a good alternative to Qwen3.7 Max?

Qwen3.5 397B A17B (78.8%) is the closest alternative on this metric, followed by Qwen3.7 Plus (78.0%). See the full ranking above for the tradeoffs.

How many Qwen models are there?

modelgrep tracks 49 Qwen models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Qwen3.7 Max. 21 of them qualify for this ranking.