Model comparison

Mellum2-12B-A2.5B-Thinking vs ZAYA1-8B

Data verified July 23, 2026

Head-to-head evidence from 4 shared benchmark results across 3 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

Mellum2-12B-A2.5B-Thinking

JetBrains

N/A

No comparison

ZAYA1-8B

Zyphra

N/A

1 category wins1 category wins

Evidence parity. Mellum2-12B-A2.5B-Thinking and ZAYA1-8B share 4 comparable benchmark results. 2 of 8 categories are comparable. 1 result is unique to Mellum2-12B-A2.5B-Thinking; 7 to ZAYA1-8B.

Updated July 23, 2026

Shared results: 4
Mellum2-12B-A2.5B-Thinking only: 1
ZAYA1-8B only: 7
Comparable categories: 2 / 8

Treat this as a split decision. Mellum2-12B-A2.5B-Thinking makes more sense if instruction following is the priority; ZAYA1-8B is the better fit if knowledge is the priority or you need the larger 131K context window.

Confidence note. This is a partial-evidence comparison with 4 shared benchmark results across 3 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Mellum2-12B-A2.5B-Thinking and ZAYA1-8B finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.

ZAYA1-8B gives you the larger context window at 131K, compared with 128K for Mellum2-12B-A2.5B-Thinking.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Mellum2-12B-A2.5B-Thinking and ZAYA1-8B
Category	Mellum2-12B-A2.5B-Thinking	Δ	ZAYA1-8B
Knowledge	Mellum2-12B-A2.5B-Thinking57.6	Margin→ 16.0	ZAYA1-8B73.6
Inst. Following	Mellum2-12B-A2.5B-Thinking76.5	Margin← 12.4	ZAYA1-8B64.1
Math	Mellum2-12B-A2.5B-ThinkingNot measured	MarginNo overlap	ZAYA1-8B80.4

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

A · Mellum2-12B-A2.5B-ThinkingB · ZAYA1-8B

GPQA
Knowledge
Source ↗
A 57.6%B 71%
Winner: ZAYA1-8BΔ 13.4
GPQA: Mellum2-12B-A2.5B-Thinking scored 57.6%; ZAYA1-8B scored 71%. ZAYA1-8B wins this benchmark.
IFEval
Inst. Following
Source ↗
A 76.5%B 85.6%
Winner: ZAYA1-8BΔ 9.1
IFEval: Mellum2-12B-A2.5B-Thinking scored 76.5%; ZAYA1-8B scored 85.6%. ZAYA1-8B wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

Metric	Mellum2-12B-A2.5B-Thinking	ZAYA1-8B	Comparison
Input / output priceUSD per 1M tokens	Mellum2-12B-A2.5B-ThinkingNot available	ZAYA1-8B$0 input / $0 output	A complete price comparison is not available.
Generation speedtokens per second	Mellum2-12B-A2.5B-ThinkingNot available	ZAYA1-8BNot available	A complete speed comparison is not available.
First-answer latencyseconds to first token	Mellum2-12B-A2.5B-ThinkingNot available	ZAYA1-8BNot available	A complete latency comparison is not available.
Context windowmaximum listed tokens	Mellum2-12B-A2.5B-Thinking128K	ZAYA1-8B131K	ZAYA1-8B lists the larger context window.

Benchmark Deep Dive

Agentic

1 benchmarks

Benchmark	Mellum2-12B-A2.5B-Thinking	ZAYA1-8B	Result
BFCL v4Source	45.6%	39.2%	Mellum2-12B-A2.5B-Thinking leads

Coding

1 benchmarks

Benchmark	Mellum2-12B-A2.5B-Thinking	ZAYA1-8B	Result
LiveCodeBench v6Source	—	65.8%	Not comparable

KnowledgeZAYA1-8B wins

4 benchmarks

Benchmark	Mellum2-12B-A2.5B-Thinking	ZAYA1-8B	Result
MMLU-ReduxSource	86.2%	—	Not comparable
GPQASource	57.6%	71%	ZAYA1-8B leads
GPQA-DSource	57.6%	71.0%	ZAYA1-8B leads
MMLU-ProSource	—	74.2%	Not comparable

Math

4 benchmarks

Benchmark	Mellum2-12B-A2.5B-Thinking	ZAYA1-8B	Result
AIME26Source	—	89.1%	Not comparable
HMMT Feb 2026Source	—	71.6%	Not comparable
IMOAnswerBenchSource	—	59.3%	Not comparable
ApexSource	—	32.2%	Not comparable

Inst. FollowingMellum2-12B-A2.5B-Thinking wins

2 benchmarks

Benchmark	Mellum2-12B-A2.5B-Thinking	ZAYA1-8B	Result
IFEvalSource	76.5%	85.6%	ZAYA1-8B leads
IFBenchSource	—	52.6%	Not comparable

Frequently Asked Questions (3)

Which is better, Mellum2-12B-A2.5B-Thinking or ZAYA1-8B?

Mellum2-12B-A2.5B-Thinking and ZAYA1-8B are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.

Which is better for knowledge tasks, Mellum2-12B-A2.5B-Thinking or ZAYA1-8B?

ZAYA1-8B has the edge for knowledge tasks in this comparison, averaging 73.6 versus 57.6. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for instruction following, Mellum2-12B-A2.5B-Thinking or ZAYA1-8B?

Mellum2-12B-A2.5B-Thinking has the edge for instruction following in this comparison, averaging 76.5 versus 64.1. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

JetBrains Compare Pricing Methodology Find Your Best LLM Overall Rankings

Last updated: July 23, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.