Skip to main content
BenchLM
Advertise

Model comparison

Inkling-Small vs Qwen3.8 Max

Updated August 3, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Inkling-Small

Thinking Machines Lab

59.9/100

Estimated · Public rank #51

90% interval 50.0–69.8

Qwen3.8 Max

Alibaba

65.4/100

Estimated · Public rank #31

90% interval 55.5–75.3

Qwen3.8 Max has the higher public score estimate, 65.4 versus 59.91, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

10 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

No workload recommendation clears the current evidence threshold.

Use the matched evidence, workload costs, and sourced specifications below instead of treating a point score as a universal answer.

Show secondary and unsupported calls

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

Shared results
10
Inkling-Small only
10
Qwen3.8 Max only
42
Like-for-like categories
3 / 8

1 category uses different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Knowledge

Like-for-like
Inkling-Small
53.4
Qwen3.8 Max
50.2
Weighted basis
2 vs 2 rows
Reading
Inkling-Small leads

Multimodal

Like-for-like
Inkling-Small
76.6
Qwen3.8 Max
86.3
Weighted basis
2 vs 2 rows
Reading
Qwen3.8 Max leads

Instruction following

Like-for-like
Inkling-Small
82.2
Qwen3.8 Max
82.8
Weighted basis
1 vs 1 rows
Reading
Qwen3.8 Max leads

Coding

Directional only
Inkling-Small
62.4
Qwen3.8 Max
67.7
Weighted basis
3 vs 1 rows
Reading
Directional only

Agentic

Not comparable
Inkling-Small
70.1
Qwen3.8 Max
86.1
Weighted basis
2 vs 1 rows
Reading
Not comparable

Reasoning

Not comparable
Inkling-Small
40.1
Qwen3.8 Max
78.3
Weighted basis
1 vs 2 rows
Reading
Not comparable

Math

Not comparable
Inkling-Small
92.9
Qwen3.8 Max
Not measured
Weighted basis
2 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
Inkling-Small
Not measured
Qwen3.8 Max
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Inkling-Small
$0.0013
Fits in one request
Qwen3.8 Max
API rate not published
Fits in one request

Qwen3.8 Max has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Inkling-Small
$0.03332
Fits in one request
Qwen3.8 Max
API rate not published
Fits in one request

Qwen3.8 Max has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Inkling-Small
$0.0492
Fits in one request
Qwen3.8 Max
API rate not published
Fits in one request
Cached-input rate unavailable

Qwen3.8 Max has no comparable published API token rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Inkling-Small

$0.116 per 1M cached input tokens

Qwen3.8 Max

No comparable hosted API rate

Alibaba Cloud Model Studio pricing

Documented inputs

Inkling-Small

Not sourced

Qwen3.8 Max

Not sourced

Documented outputs

Inkling-Small

Not sourced

Qwen3.8 Max

Not sourced

Provider availability

Inkling-Small

Not sourced

Qwen3.8 Max

Not sourced

Reasoning profile

Inkling-Small

Hybrid

Qwen3.8 Max

Reasoning

Weight access

Inkling-Small

Open Weight

Qwen3.8 Max

Proprietary

License

Inkling-Small

Open Weight

Qwen3.8 Max

Proprietary

Release date

Inkling-Small

2026-07-30

Qwen3.8 Max

2026-08-03

If you already use one of these models
Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
Qwen3.8 Max has the higher public score estimate, 65.4 versus 59.91, but the 90% score intervals overlap.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Both models list 1M.

Run the same representative tasks against both endpoints before changing production traffic.

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence62 rows

Agentic

  • Terminal-Bench 2.0

    Inkling-Small64.7%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • BrowseComp

    Inkling-Small77.4%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • MCP Atlas

    Inkling-Small79.6%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • Toolathlon-Verified

    Inkling-Small54.4%
    Source
    Qwen3.8 Max72.5%
    Source

    Qwen3.8 Max leads this result

  • Terminal-Bench 2.1

    Inkling-Small—
    Qwen3.8 Max86.6%
    Source

    Not directly comparable

  • CoWorkBench

    Inkling-Small—
    Qwen3.8 Max74.8%
    Source

    Not directly comparable

  • JobBench

    Inkling-Small—
    Qwen3.8 Max53.4%
    Source

    Not directly comparable

  • skillsBench

    Inkling-Small—
    Qwen3.8 Max70.2%
    Source

    Not directly comparable

  • Agents' Last Exam

    Inkling-Small—
    Qwen3.8 Max52.4%
    Source

    Not directly comparable

  • AutomationBench

    Inkling-Small—
    Qwen3.8 Max27.3%
    Source

    Not directly comparable

  • WideResearch

    Inkling-Small—
    Qwen3.8 Max81.9%
    Source

    Not directly comparable

  • HLE w/ tools

    Inkling-Small—
    Qwen3.8 Max56.2%
    Source

    Not directly comparable

  • OSWorld-Verified

    Inkling-Small—
    Qwen3.8 Max86.1%
    Source

    Not directly comparable

  • OSWorld 2.0

    Inkling-Small—
    Qwen3.8 Max19.4%
    Source

    Not directly comparable

  • WebArena-Verified

    Inkling-Small—
    Qwen3.8 Max66.8%
    Source

    Not directly comparable

  • AndroidWorld

    Inkling-Small—
    Qwen3.8 Max85.3%
    Source

    Not directly comparable

  • MobileWorld

    Inkling-Small—
    Qwen3.8 Max77.8%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    Inkling-Small80.2%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • SWE-bench Pro

    Inkling-Small55.9%
    Source
    Qwen3.8 Max67.7%
    Source

    Qwen3.8 Max leads this result

  • Terminal-Bench 2.0

    Inkling-Small64.7%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • SciCode

    Inkling-Small48.7%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • Terminal-Bench 2.1

    Inkling-Small—
    Qwen3.8 Max86.6%
    Source

    Not directly comparable

  • deepSwe

    Inkling-Small—
    Qwen3.8 Max56.6%
    Source

    Not directly comparable

  • NL2Repo

    Inkling-Small—
    Qwen3.8 Max55.9%
    Source

    Not directly comparable

  • FrontierSWE

    Inkling-Small—
    Qwen3.8 Max73.5%
    Source

    Not directly comparable

  • MLS-Bench Lite

    Inkling-Small—
    Qwen3.8 Max41.0%
    Source

    Not directly comparable

  • PaperBench

    Inkling-Small—
    Qwen3.8 Max93.0%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Inkling-Small40.1%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • CritPt

    Inkling-Small8.3%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • MRCRv2

    Inkling-Small—
    Qwen3.8 Max92.9%
    Source

    Not directly comparable

  • LongBench v2

    Inkling-Small—
    Qwen3.8 Max66.3%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Inkling-Small89.5%
    Source
    Qwen3.8 Max92.6%
    Source

    Qwen3.8 Max leads this result

  • GPQA-D

    Inkling-Small89.5%
    Source
    Qwen3.8 Max92.6%
    Source

    Qwen3.8 Max leads this result

  • HLE

    Inkling-Small47.8%
    Source
    Qwen3.8 Max43.6%
    Source

    Inkling-Small leads this result

  • HLE w/o tools

    Inkling-Small31.6%
    Source
    Qwen3.8 Max43.6%
    Source

    Qwen3.8 Max leads this result

Math

  • AIME26

    Inkling-Small95.5%
    Source
    Qwen3.8 Max—

    Not directly comparable

  • HMMT Feb 2026

    Inkling-Small90.2%
    Source
    Qwen3.8 Max—

    Not directly comparable

Multimodal

  • MMMU-Pro

    Inkling-Small74%
    Source
    Qwen3.8 Max82.3%
    Source

    Qwen3.8 Max leads this result

  • CharXiv

    Inkling-Small81.3%
    Source
    Qwen3.8 Max93.5%
    Source

    Qwen3.8 Max leads this result

  • CharXiv w/o tools

    Inkling-Small77.4%
    Source
    Qwen3.8 Max88.4%
    Source

    Qwen3.8 Max leads this result

  • MathVision

    Inkling-Small—
    Qwen3.8 Max95.2%
    Source

    Not directly comparable

  • MathVision w/ Python

    Inkling-Small—
    Qwen3.8 Max97.7%
    Source

    Not directly comparable

  • BabyVision

    Inkling-Small—
    Qwen3.8 Max82.0%
    Source

    Not directly comparable

  • BabyVision w/ Python

    Inkling-Small—
    Qwen3.8 Max91.3%
    Source

    Not directly comparable

  • ZeroBench

    Inkling-Small—
    Qwen3.8 Max24.0%
    Source

    Not directly comparable

  • ZeroBench w/ Python

    Inkling-Small—
    Qwen3.8 Max49.0%
    Source

    Not directly comparable

  • MedXpertQA (MM)

    Inkling-Small—
    Qwen3.8 Max80.4%
    Source

    Not directly comparable

  • ScreenSpot Pro

    Inkling-Small—
    Qwen3.8 Max84.5%
    Source

    Not directly comparable

  • Vision2Web

    Inkling-Small—
    Qwen3.8 Max69.0%
    Source

    Not directly comparable

  • OmniDocBench 1.5

    Inkling-Small—
    Qwen3.8 Max92.1%
    Source

    Not directly comparable

  • OCRBench V2

    Inkling-Small—
    Qwen3.8 Max74.2%
    Source

    Not directly comparable

  • CC-OCR

    Inkling-Small—
    Qwen3.8 Max79.6%
    Source

    Not directly comparable

  • RealWorldQA

    Inkling-Small—
    Qwen3.8 Max88.0%
    Source

    Not directly comparable

  • ERQA

    Inkling-Small—
    Qwen3.8 Max77.8%
    Source

    Not directly comparable

  • SimpleVQA

    Inkling-Small—
    Qwen3.8 Max75.0%
    Source

    Not directly comparable

  • PerceptionBench

    Inkling-Small—
    Qwen3.8 Max63.5%
    Source

    Not directly comparable

  • Video-MME (with subtitle)

    Inkling-Small—
    Qwen3.8 Max90.4%
    Source

    Not directly comparable

  • VideoMMMU

    Inkling-Small—
    Qwen3.8 Max88.7%
    Source

    Not directly comparable

  • MMVU

    Inkling-Small—
    Qwen3.8 Max82.4%
    Source

    Not directly comparable

  • MLVU (M-Avg)

    Inkling-Small—
    Qwen3.8 Max90.8%
    Source

    Not directly comparable

  • LVBench

    Inkling-Small—
    Qwen3.8 Max81.8%
    Source

    Not directly comparable

Instruction following

  • IFBench

    Inkling-Small82.2%
    Source
    Qwen3.8 Max82.8%
    Source

    Qwen3.8 Max leads this result

Frequently asked questions

Which is better, Inkling-Small or Qwen3.8 Max?

Qwen3.8 Max has the higher public score estimate, 65.4 versus 59.91, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Inkling-Small or Qwen3.8 Max?

The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Inkling-Small or Qwen3.8 Max?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, Inkling-Small or Qwen3.8 Max?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Inkling-Small or Qwen3.8 Max?

Both models list the same context window, 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 3, 2026

Watch Inkling-Small vs Qwen3.8 Max

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.