Interlinked.

Methodology -- April 2026

The benchmark
trap.

The #1 coding model might not actually be #1. Published benchmarks don't normalize for thinking mode: some models are tested with reasoning on, others with it off, and one model gets worse at coding when thinking is enabled.

MiniMax M2.580.2% (always ON)Kimi K2.576.8% (thinking OFF)GLM-577.8% (always ON)Qwen 3.576.4% (ON)DeepSeek V3.267.8% (thinking OFF)FAIR SCOREHANDICAPPED

Models with fair benchmarks

4/7

same mode published as benchmarked

Misleading comparisons

3/7

handicapped, inflated, or mode-switching

The trap

Three ways thinking mode breaks comparisons.

Always-on thinking inflates scores. MiniMax M2.5 runs interleaved thinking on every token: baked in, no off switch. Comparing its 80.2% SWE-bench to models tested without thinking isn't a model comparison. It's a methodology comparison.

Some models are tested with thinking off. Kimi K2.5 was benchmarked in instant mode (thinking explicitly disabled) because their eval harness was incompatible. Their 76.8% SWE-bench is their floor, not their ceiling.

Thinking can hurt. DeepSeek V3.2 is worse at coding with thinking on: 50.7% vs 56.1% on LiveCodeBench. But thinking helps its math. DeepSeek switches modes between benchmarks.

Handicapped

2 of 7

Kimi K2.5, DeepSeek V3.2

Unfair advantage

1 of 7

MiniMax M2.5 (always on)

Thinking hurts code

-5.4 pts

DeepSeek V3.2 on LiveCodeBench

Full audit

Thinking mode status for every major model.

ModelThinking modeWhat was publishedImpact
Claude Opus 4.6Adaptive (always on)Auto-adjusts reasoning depth. No off switchFair
MiniMax M2.5Always on (architectural)Interleaved thinking baked in. Cannot be disabledUnfair advantage vs models tested without thinking
Kimi K2.5Toggle (instant vs thinking)Benchmarked in INSTANT mode: thinking OFF for SWE-bench, AIME, GPQAHandicapped. Real scores unknown
GLM-5Always onReasoning integrated into architectureFair
DeepSeek V3.2ToggleThinking HURTS coding: 50.7% vs 56.1% non-thinking on LCBMode-switching between benchmarks
Gemma 4 26B / 31BToggle (think: true)Thinking ON for all published benchmarksFair
Qwen 3.5ToggleThinking ON for published benchmarksFair

Highlighted rows are the problematic ones. The thinking mode status changes the interpretation of every published score.

SWE-bench annotated

The coding leaderboard nobody publishes.

Same scores every leaderboard shows. Different column: whether thinking was on.

ModelSWE-benchThinking?What this means
MiniMax M2.580.2%Always ONFair score, but unfair comparison
GLM-577.8%Always ONFair score
Kimi K2.576.8%OFF (instant)3.4% gap with MiniMax might not exist
Qwen 3.5 397B76.4%ONFair score
Qwen3-Coder-Next70.6%ON80B MoE, 3B active
DeepSeek V3.267.8%OFF (for code)Thinking hurts its coding. Non-thinking intentional

If Kimi K2.5's thinking-on SWE-bench lands above 80%, the entire leaderboard reshuffles.

Their eval harness couldn't handle thinking mode output at publication time.

The Kimi question

What if Kimi K2.5 had thinking on?

Kimi scores 87.6% on GPQA with thinking off. MiniMax scores 47% with thinking always on. That's a 40-point gap in the wrong direction: the handicapped model crushes the unhandicapped one on science.

On math, Kimi hits 96.1% on AIME with thinking off: the highest score of any model, period. MiniMax estimates sit at 78-88%.

The implication: Kimi K2.5 with thinking on might be the best all-around model in the open-weight world. We just don't have the numbers.

BenchmarkKimi (OFF)MiniMax (ON)Kimi (ON)
SWE-bench76.8%80.2%Unknown
AIME 202696.1%~78-88%Unknown
GPQA Diamond87.6%47%Unknown
LiveCodeBench85.0%UnknownUnknown

GPQA Diamond -- Kimi (off) vs MiniMax (on)

87.6%

Kimi K2.5 (thinking OFF)

vs
47%

MiniMax M2.5 (thinking ON)

The handicapped model crushes the unhandicapped one by 40 points on science. The leaderboard ranking depends entirely on which benchmark you look at.

The DeepSeek paradox

When thinking makes code worse.

DeepSeek publishes coding scores with thinking off and math scores with thinking on, switching modes between benchmarks.

BenchmarkNon-thinkingThinkingDifference
LiveCodeBench v656.1%50.7%-5.4 pts
AIME 2026 (math)83.5%86.3%+2.8 pts
ImplicationPublished coding scoresPublished math scoresDifferent mode per benchmark

What to do about it

How to read benchmarks honestly.

Rule 1

Always check the thinking mode status. Model cards and paper appendices sometimes disclose this. If they don't, assume the benchmark used whatever mode produces the highest score.

Rule 2

Compare models at the same thinking mode. MiniMax M2.5 (always on) vs Kimi K2.5 (thinking off) isn't a fair fight. Wait for normalized scores, or discount the always-on model's advantage.

Rule 3

The scaffold matters more than the model. SWE-bench Pro shows a 22+ point swing between different agent scaffolds using the same model. That's 3x the gap between the best and worst models on this leaderboard.

The bigger picture

Context is the multiplier, not the model.

The model is one variable. The retrieval infrastructure (how context is retrieved, how tasks are decomposed, how code is navigated) matters more than which model sits behind it.

A 22-point scaffold swing on SWE-bench Pro vs a 3-point model gap. The thinking mode debate matters for benchmark integrity, but in production, the context window is the bottleneck, not the reasoning chain.

That's where Interlinked Files lives. Machine-wide retrieval that turns any model into a better coding agent, regardless of whether thinking is on or off.

Thinking mode status verified from official model cards, release announcements, API documentation, and community testing. Benchmark scores from published papers and Arena Leaderboard. All data as of April 2026.