Skip to main content

Model comparison

Inkling vs ZAYA1-74B-Preview

Data verified

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

Thinking Machines Lab
69/100
Margin
14.0pts
← winning
55/100
2 category wins1 category wins

Verified leaderboard positions: Inkling #17; ZAYA1-74B-Preview unranked

Evidence parity. Inkling and ZAYA1-74B-Preview share 4 comparable benchmark results. 3 of 8 categories are comparable. 12 results are unique to Inkling; 3 to ZAYA1-74B-Preview.

Updated July 15, 2026
Shared results
4
Inkling only
12
ZAYA1-74B-Preview only
3
Comparable categories
3 / 8

Pick Inkling if you want the stronger benchmark profile. ZAYA1-74B-Preview only becomes the better choice if knowledge is the priority or you want the cheaper token bill.

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

Why this result

Inkling is clearly ahead on the provisional aggregate, 69 to 55. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Inkling's sharpest advantage is in mathematics, where it averages 97.1 against 76.4. The single biggest benchmark swing on the page is GPQA, 87.9% to 57.3%. ZAYA1-74B-Preview does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.

Inkling is also the more expensive model on tokens at $1.87 input / $4.68 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for ZAYA1-74B-Preview. That is roughly Infinityx on output cost alone. ZAYA1-74B-Preview is the reasoning model in the pair, while Inkling is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Inkling gives you the larger context window at 1M, compared with 256K for ZAYA1-74B-Preview.

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 Inkling and ZAYA1-74B-Preview
CategoryInklingΔZAYA1-74B-Preview
MathInkling97.1Margin 20.7ZAYA1-74B-Preview76.4
CodingInkling68.6Margin 15.4ZAYA1-74B-Preview53.2
KnowledgeInkling51.7Margin 14.4ZAYA1-74B-Preview66.1
AgenticInkling69.4MarginNo overlapZAYA1-74B-PreviewNot measured
MultimodalInkling76.5MarginNo overlapZAYA1-74B-PreviewNot measured
Inst. FollowingInkling79.8MarginNo overlapZAYA1-74B-PreviewNot measured

Decisive benchmark drivers

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

More
A · InklingB · ZAYA1-74B-Preview
  1. GPQA

    Knowledge
    Source ↗
    A 87.9%B 57.3%
    Winner: InklingΔ 30.6
    GPQA: Inkling scored 87.9%; ZAYA1-74B-Preview scored 57.3%. Inkling wins this benchmark.
  2. SWE-bench Verified

    Coding
    Source ↗
    A 77.6%B 53.2%
    Winner: InklingΔ 24.4
    SWE-bench Verified: Inkling scored 77.6%; ZAYA1-74B-Preview scored 53.2%. Inkling wins this benchmark.
  3. AIME26

    Math
    Source ↗
    A 97.1%B 76.4%
    Winner: InklingΔ 20.7
    AIME26: Inkling scored 97.1%; ZAYA1-74B-Preview scored 76.4%. Inkling wins this benchmark.

Operational comparison

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

MetricInklingZAYA1-74B-PreviewComparison
Input / output priceUSD per 1M tokensInkling$1.87 input / $4.68 outputZAYA1-74B-Preview$0 input / $0 outputZAYA1-74B-Preview has the lower combined listed price.
Generation speedtokens per secondInklingNot availableZAYA1-74B-PreviewNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenInklingNot availableZAYA1-74B-PreviewNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensInkling1MZAYA1-74B-Preview256KInkling lists the larger context window.

Benchmark Deep Dive

Agentic
BenchmarkInklingZAYA1-74B-PreviewResult
Terminal-Bench 2.0Source 63.8%Not comparable
BrowseCompSource 77.1%Not comparable
MCP AtlasSource 74.1%Not comparable
Design Arena Agentic Web DevSource 1258Not comparable
τ²-bench AirlineSource 56.1%Not comparable
CodingInkling wins
BenchmarkInklingZAYA1-74B-PreviewResult
SWE-bench VerifiedSource 77.6%53.2%Inkling leads
SWE-bench ProSource 54.3%Not comparable
Terminal-Bench 2.0Source 63.8%Not comparable
LiveCodeBench v6Source 65.7%Not comparable
KnowledgeZAYA1-74B-Preview wins
BenchmarkInklingZAYA1-74B-PreviewResult
GPQASource 87.9%57.3%Inkling leads
GPQA-DSource 87.9%57.3%Inkling leads
HLESource 46%Not comparable
HLE w/o toolsSource 30%Not comparable
MMLU-ProSource 68.1%Not comparable
MathInkling wins
BenchmarkInklingZAYA1-74B-PreviewResult
AIME26Source 97.1%76.4%Inkling leads
Multimodal
BenchmarkInklingZAYA1-74B-PreviewResult
MMMU-ProSource 73.5%Not comparable
CharXivSource 82%Not comparable
CharXiv w/o toolsSource 78.1%Not comparable
Inst. Following
BenchmarkInklingZAYA1-74B-PreviewResult
IFBenchSource 79.8%Not comparable
Frequently Asked Questions (4)

Which is better, Inkling or ZAYA1-74B-Preview?

Inkling is ahead on BenchLM's provisional leaderboard, 69 to 55. The biggest single separator in this matchup is GPQA, where the scores are 87.9% and 57.3%.

Which is better for knowledge tasks, Inkling or ZAYA1-74B-Preview?

ZAYA1-74B-Preview has the edge for knowledge tasks in this comparison, averaging 66.1 versus 51.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, Inkling or ZAYA1-74B-Preview?

Inkling has the edge for coding in this comparison, averaging 68.6 versus 53.2. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for math, Inkling or ZAYA1-74B-Preview?

Inkling has the edge for math in this comparison, averaging 97.1 versus 76.4. Inside this category, AIME26 is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 15, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.