Skip to main content
BenchLM
Data

Inkling 2 Small vs Step 5 Preview

Updated November 7, 2026. Rank says Inkling 2 Small is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

Decision reading

Inkling 2 Small has the higher public point estimate, 59.06 versus 56.83. Their conditional score ranges overlap. These ranges do not establish rank confidence. 8 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A
Thinking Machines Lab logo
Inkling 2 Small

Thinking Machines Lab

59.06/100

Estimated · Public rank #56

Conditional range 44.7–73.4

Model B
StepFun logo
Step 5 Preview

StepFun

56.83/100

Estimated · Public rank #61

Conditional range 47.1–66.5

Shared results
8
Inkling 2 Small only
17
Step 5 Preview only
16
Like-for-like categories
0 / 8
Estimated: Inkling 2 Small and Step 5 Preview. Conditional ranges do not establish rank confidence.How the comparison works

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.

  • Chat turn cost

    1K fresh input + 500 output tokens

    Inkling 2 Small

    Inkling 2 Small has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

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

    Inkling 2 Small

    Inkling 2 Small has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    Inkling 2 Small

    Inkling 2 Small has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    Step 5 Preview is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    Step 5 Preview is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

59.7Inkling 2 Small49.2Step 5 Preview

Directional only · BenchAlign v5.8

Inkling 2 Small has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Open the coding leaderboardShortlist models for this job
Same basis rules as the category table below

What is actually comparable

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

4 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

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.

Bars run 0–100 on each benchmark’s normalized display scale

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Directional only
Inkling 2 Small
59.9
Supported · #26/131
Step 5 Preview
54.8
Estimated · #43/131
Basis
BenchAlign v5.8 lane · 5 vs 13 public rows
Reading
Directional only

Coding

Directional only
Inkling 2 Small
59.7
Supported · #20/157
Step 5 Preview
49.2
Estimated · #48/157
Basis
BenchAlign v5.8 lane · 6 vs 6 public rows
Reading
Directional only

Multimodal

Directional only
Inkling 2 Small
86.6
#12/59
Step 5 Preview
72.3
#34/59
Basis
Provisional lane · 1 vs 2 weighted rows
Reading
Directional only

Knowledge

Directional only
Inkling 2 Small
60.4
Estimated · #46/178
Step 5 Preview
61.0
Supported · #44/178
Basis
BenchAlign v5.8 lane · 2 vs 2 public rows
Reading
Directional only

Reasoning

Not comparable
Inkling 2 Small
86.1
Unranked · 2 rankable rows
Step 5 Preview
80.8
Unranked · 2 rankable rows
Basis
Provisional lane · 1 vs 0 weighted rows
Reading
Not comparable

Multilingual

Not comparable
Inkling 2 Small
Not ranked
Step 5 Preview
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
Inkling 2 Small
Not ranked
Step 5 Preview
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
Inkling 2 Small
85.9
Unranked · 2 rankable rows
Step 5 Preview
Not ranked
Basis
Provisional lane · 1 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

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 2 Small
$0.0008
Fits in one request
Step 5 Preview
$0.00235
Fits in one request

Inkling 2 Small has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Inkling 2 Small
$0.018
Fits in one request
Step 5 Preview
$0.0581
Fits in one request

Inkling 2 Small has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

Inkling 2 Small
$0.022
Fits in one request
Step 5 Preview
$0.057
Fits in one request

Inkling 2 Small has the lower modeled cost

Costs use the listed standard API rates.

Cached input falls back to the list input rate only where a cached rate is unpublished

Specification differences

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

Reasoning profile

Inkling 2 Small

Hybrid

Step 5 Preview

Reasoning

Weight access

Inkling 2 Small

Proprietary

Step 5 Preview

Open Weight

License

Inkling 2 Small

Proprietary

Step 5 Preview

Open Weight

Release date

Inkling 2 Small

2026-11-05

Step 5 Preview

2026-09-20

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
Inkling 2 Small has the higher public point estimate, 59.06 versus 56.83. Their conditional score ranges overlap. These ranges do not establish rank confidence.
Workload cost
Repository review: $0.018 vs $0.0581. Cache-heavy agent loop: $0.022 vs $0.057.
Context tradeoff
Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Inkling 2 Small or Step 5 Preview?

Inkling 2 Small has the higher public point estimate, 59.06 versus 56.83. Their conditional score ranges overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, Inkling 2 Small or Step 5 Preview?

Inkling 2 Small scores higher for coding on the public lane, 59.7 to 49.2. Step 5 Preview is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Inkling 2 Small or Step 5 Preview?

Inkling 2 Small scores higher for agentic tasks on the public lane, 59.9 to 54.8. Step 5 Preview is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Inkling 2 Small or Step 5 Preview?

For the stated presets, chat costs $0.0008 on Inkling 2 Small and $0.00235 on Step 5 Preview; repository review costs $0.018 and $0.0581; the cache-heavy agent loop costs $0.022 and $0.057. Costs use the listed standard API rates.

Which has the larger context window, Inkling 2 Small or Step 5 Preview?

Both models list the same context window, 1M.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

Inkling 2 Small
API / mo$975
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Step 5 Preview
API / mo$2,775
Self-host / mo$27,962
Break-even617M/day
Model the full break-even

Benchmark evidence

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

Browse raw public benchmark evidence41 rows

Agentic

  • Terminal-Bench 2.1

    Inkling 2 Small85.1%
    Source
    Step 5 Preview85.0%
    Source

    Inkling 2 Small leads this result

  • BrowseComp

    Inkling 2 Small90.2%
    Source
    Step 5 Preview88.7%
    Source

    Inkling 2 Small leads this result

  • DeepSearchQA

    Inkling 2 Small90.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • MCP Atlas

    Inkling 2 Small85%
    Source
    Step 5 Preview85.6%
    Source

    Step 5 Preview leads this result

  • JobBench

    Inkling 2 Small70.5%
    Source
    Step 5 Preview59.0%
    Source

    Inkling 2 Small leads this result

  • Terminal-Bench 4.0

    Inkling 2 Small—
    Step 5 Preview33.30%
    Source

    Not directly comparable

  • CyberGym

    Inkling 2 Small—
    Step 5 Preview84.7%
    Source

    Not directly comparable

  • AutomationBench

    Inkling 2 Small—
    Step 5 Preview44.0%
    Source

    Not directly comparable

  • Toolathlon-Verified

    Inkling 2 Small—
    Step 5 Preview74.1%
    Source

    Not directly comparable

  • APEX-Agents

    Inkling 2 Small—
    Step 5 Preview37.8%
    Source

    Not directly comparable

  • DRACO

    Inkling 2 Small—
    Step 5 Preview83.3%
    Source

    Not directly comparable

  • Agents' Last Exam

    Inkling 2 Small—
    Step 5 Preview29.5%
    Source

    Not directly comparable

  • Hermes Index

    Inkling 2 Small—
    Step 5 Preview32.8%
    Source

    Not directly comparable

  • Hermes Bench

    Inkling 2 Small—
    Step 5 Preview69.2%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    Inkling 2 Small82.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • SWE Multilingual

    Inkling 2 Small81.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • sweMarathon

    Inkling 2 Small38.8%
    Source
    Step 5 Preview72.7%
    Source

    Step 5 Preview leads this result

  • Terminal-Bench 2.1

    Inkling 2 Small85.1%
    Source
    Step 5 Preview85.0%
    Source

    Inkling 2 Small leads this result

  • DeepSWE

    Inkling 2 Small72.1%
    Source
    Step 5 Preview67.7%
    Source

    Inkling 2 Small leads this result

  • PostTrainBench v1.2

    Inkling 2 Small31.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • SciCode

    Inkling 2 Small—
    Step 5 Preview58.9%
    Source

    Not directly comparable

  • ProgramBench

    Inkling 2 Small—
    Step 5 Preview80.5%
    Source

    Not directly comparable

  • MLS-Bench Lite

    Inkling 2 Small—
    Step 5 Preview40.5%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    Inkling 2 Small70.6%
    Source
    Step 5 Preview—

    Not directly comparable

  • CritPt

    Inkling 2 Small—
    Step 5 Preview20.9%
    Source

    Not directly comparable

Multimodal

  • CharXiv

    Inkling 2 Small87.7%
    Source
    Step 5 Preview—

    Not directly comparable

  • CharXiv w/o tools

    Inkling 2 Small85.6%
    Source
    Step 5 Preview—

    Not directly comparable

  • Chartography (no tools)

    Inkling 2 Small36.7%
    Source
    Step 5 Preview—

    Not directly comparable

  • Chartography (tools)

    Inkling 2 Small80.4%
    Source
    Step 5 Preview—

    Not directly comparable

  • BabyVision

    Inkling 2 Small68.6%
    Source
    Step 5 Preview—

    Not directly comparable

  • BabyVision w/ Python

    Inkling 2 Small93.7%
    Source
    Step 5 Preview—

    Not directly comparable

  • ScreenSpot Pro

    Inkling 2 Small83.8%
    Source
    Step 5 Preview—

    Not directly comparable

  • PerceptionBench

    Inkling 2 Small64.7%
    Source
    Step 5 Preview—

    Not directly comparable

  • MathVision

    Inkling 2 Small92.8%
    Source
    Step 5 Preview—

    Not directly comparable

  • OfficeQA Pro

    Inkling 2 Small—
    Step 5 Preview60.3%
    Source

    Not directly comparable

  • MMMU-Pro

    Inkling 2 Small—
    Step 5 Preview76%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Inkling 2 Small93%
    Source
    Step 5 Preview—

    Not directly comparable

  • GPQA-D

    Inkling 2 Small93.0%
    Source
    Step 5 Preview93.5%
    Source

    Step 5 Preview leads this result

  • HLE

    Inkling 2 Small—
    Step 5 Preview46.5%
    Source

    Not directly comparable

Math

  • HMMT Feb 2026

    Inkling 2 Small95.8%
    Source
    Step 5 Preview—

    Not directly comparable

  • Apex

    Inkling 2 Small63.5%
    Source
    Step 5 Preview—

    Not directly comparable

41 public results · 8 shared

Related comparisons

Compare API pricingRead the methodologyOpen the model selector

Watch Inkling 2 Small vs Step 5 Preview

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

Read a sample issue

Join 6,100+ subscribers.

Last updated November 7, 2026