← Back to Leaderboard

Q
Qwen | Qwen3 235b A22b 2507

Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following, logical reasoning, math, code, and tool usage. The model supports a native 262K context length and does not implement "thinking mode" (<think> blocks). Compared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench.

Compare

Fair

49.0
Most Recent Test

Significant guardrail issues may impede work.

Strengths

  • Excellent at 1.4. Conversational AI
  • Excellent at 2.2. Universality of Sin
  • Excellent at 3.2. Historical Jesus
  • Excellent at 3.4. The Resurrection
  • Excellent at 3.5. Universal Sinfulness

Weaknesses

  • Limited task completion capability
  • Weak doctrinal alignment
  • Struggles with 1.1. Missiological Research
  • Struggles with 1.2. Evangelistic Material
  • Struggles with 1.3. Apologetics
Overall Score
49.0
Tier 1 (Task) 70%
48.6
Tier 2 (Doctrine) 20%
38.3
Tier 3 (Worldview) 10%
73.3
Performance Profile
Visual representation of performance across all evaluated categories
Model Information
Description

Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following, logical reasoning, math, code, and tool usage. The model supports a native 262K context length and does not implement "thinking mode" (<think> blocks). Compared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench.

Provider

Qwen

Model ID

qwen/qwen3-235b-a22b-2507

Tests Run

1

Actions
Compare with other modelsView full leaderboardView all articles about this model
Insights & Analysis

Qwen3 235B A22B Instruct 2507 Review

Oct 12, 2026

Categories

Category Heatmap
Performance breakdown by category - darker green indicates stronger alignment
33
1.1
33
1.2
37
1.3
83
1.4
47
1.5
57
1.6
50
1.7
40
2.1
80
2.2
20
2.3
30
2.4
20
2.5
40
2.6
25
3.1
100
3.2
38
3.3
100
3.4
100
3.5
100
3.6
Low
High
Category Breakdown (Bar Chart)
Performance across different categories
1.x = Task Capability2.x = Gospel Core3.x = Worldview Confession

Recent Tests

Recent Test Runs
DateScoreVersionTier 1 (Task)Tier 2 (Gospel)Tier 3 (Worldview)Trust Tier
10/10/202649.01.0.048.638.373.3automated