Skip to main content
BenchLM
Data

Claude Sonnet 5 vs Inkling 2 Mini

Updated November 21, 2026. Rank says Claude Sonnet 5 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

Claude Sonnet 5 has the higher public point estimate, 61.6 versus 55.9. 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
Anthropic logo

Anthropic

61.6/100

Supported · Public rank #47

90% interval 57.6–65.7

Model B
Thinking Machines Lab logo

Thinking Machines Lab

55.9/100

Estimated · Public rank #69

Conditional range 46.0–65.9

API price per 1M tokens

Claude Sonnet 5: $2 input / $10 output

Inkling 2 Mini: $0.15 input / $0.50 output

Context window

Claude Sonnet 5: 1M

Inkling 2 Mini: 1M

Released

Claude Sonnet 5: June 30, 2026

Inkling 2 Mini: November 20, 2026

Shared results
8
Claude Sonnet 5 only
24
Inkling 2 Mini only
25
Like-for-like categories
0 / 8
Supported: Claude Sonnet 5 · Estimated: Inkling 2 Mini. 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 Mini

    Inkling 2 Mini 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 Mini

    Inkling 2 Mini 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 Mini

    Inkling 2 Mini 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

    Inkling 2 Mini 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

    Inkling 2 Mini 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.

62.5Claude Sonnet 553.9Inkling 2 Mini

Directional only · BenchAlign v6.0

Claude Sonnet 5 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.

Same basis rules as the category table below

Or describe your own job

The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

What is actually comparable

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

6 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 v6.0 lane in every category, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Directional only
Claude Sonnet 5
62.8
Supported · #46/179
Inkling 2 Mini
59.1
Estimated · #61/179
Basis
BenchAlign v6.0 lane · 10 vs 8 public rows
Reading
Directional only

Coding

Directional only
Claude Sonnet 5
62.5
Supported · #44/179
Inkling 2 Mini
53.9
Estimated · #67/179
Basis
BenchAlign v6.0 lane · 11 vs 5 public rows
Reading
Directional only

Reasoning

Directional only
Claude Sonnet 5
62.5
Supported · #39/161
Inkling 2 Mini
57.8
Estimated · #58/161
Basis
BenchAlign v6.0 lane · 0 vs 2 public rows
Reading
Directional only

Multimodal

Directional only
Claude Sonnet 5
60.6
Supported · #49/119
Inkling 2 Mini
55.5
Estimated · #62/119
Basis
BenchAlign v6.0 lane · 3 vs 13 public rows
Reading
Directional only

Knowledge

Directional only
Claude Sonnet 5
54.1
Supported · #69/158
Inkling 2 Mini
51.9
Estimated · #79/158
Basis
BenchAlign v6.0 lane · 6 vs 3 public rows
Reading
Directional only

Multilingual

Directional only
Claude Sonnet 5
59.0
Estimated · #26/87
Inkling 2 Mini
54.7
Estimated · #39/87
Basis
BenchAlign v6.0 lane · 0 vs 0 public rows
Reading
Directional only

Instruction following

Not comparable
Claude Sonnet 5
58.4
Supported · #37/143
Inkling 2 Mini
Not ranked
Basis
BenchAlign v6.0 lane · 0 vs 1 public rows
Reading
Not comparable

Math

Not comparable
Claude Sonnet 5
59.3
Supported · #35/125
Inkling 2 Mini
Not ranked
Basis
BenchAlign v6.0 lane · 2 vs 1 public rows
Reading
Not comparable

Ranks count the models scored in each category’s BenchAlign v6.0 lane, so denominators differ by category. 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

Claude Sonnet 5
$0.007
Fits in one request
Inkling 2 Mini
$0.0004
Fits in one request

Inkling 2 Mini has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Claude Sonnet 5
$0.13
Fits in one request
Inkling 2 Mini
$0.009
Fits in one request

Inkling 2 Mini has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

Claude Sonnet 5
$0.18
Fits in one request
Inkling 2 Mini
$0.011
Fits in one request

Inkling 2 Mini has the lower modeled cost

Costs use the listed standard API rates.

Your workload

Enter a month of tokens to price it at both models’ listed API rates.

Claude Sonnet 5
$40.00
per month
Inkling 2 Mini
$2.50
per month

Standard listed rates, no batch or volume discounts.

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

Price against score, among all priced models

Overall score against output price per 1M tokens, one model per family. Pricing as of November 20, 2026.

Claude Sonnet 5Inkling 2 Mini

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • 12 models score at least as high as Claude Sonnet 5 for the same price or less. The nearest is Claude Sonnet 5.5 (score 77.8, also $10.00 per 1M output tokens).
  • 2 models at the same price score higher than Inkling 2 Mini. The nearest is Claude Haiku 5.5 (score 63.7, also $0.50 per 1M output tokens).

Gray dots are the other priced models. The dark step line is the efficiency frontier: for a model under it, another model costs no more and scores at least as high.

See Inkling 2 Mini on the full price vs performance chart

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.

Claude Sonnet 5

$0.2 per 1M cached input tokens

Claude API pricing

Inkling 2 Mini

$0.015 per 1M cached input tokens

Tinker models and pricing

Reasoning profile

Claude Sonnet 5

Reasoning

Inkling 2 Mini

Hybrid

Weight access

Claude Sonnet 5

Proprietary

Inkling 2 Mini

Open Weight

License

Claude Sonnet 5

Proprietary

Inkling 2 Mini

Open Weight

Release date

Claude Sonnet 5

2026-06-30

Inkling 2 Mini

2026-11-20

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
Claude Sonnet 5 has the higher public point estimate, 61.6 versus 55.9. Their conditional score ranges overlap. These ranges do not establish rank confidence.
Workload cost
Repository review: $0.13 vs $0.009. Cache-heavy agent loop: $0.18 vs $0.011.
Context tradeoff
Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Claude Sonnet 5 or Inkling 2 Mini?

Claude Sonnet 5 has the higher public point estimate, 61.6 versus 55.9. 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, Claude Sonnet 5 or Inkling 2 Mini?

Claude Sonnet 5 scores higher for coding on the public lane, 62.5 to 53.9. Inkling 2 Mini 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, Claude Sonnet 5 or Inkling 2 Mini?

Claude Sonnet 5 scores higher for agentic tasks on the public lane, 62.8 to 59.1. Inkling 2 Mini 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, Claude Sonnet 5 or Inkling 2 Mini?

For the stated presets, chat costs $0.007 on Claude Sonnet 5 and $0.0004 on Inkling 2 Mini; repository review costs $0.13 and $0.009; the cache-heavy agent loop costs $0.18 and $0.011. Costs use the listed standard API rates.

Which has the larger context window, Claude Sonnet 5 or Inkling 2 Mini?

Both models list the same context window, 1M.

Benchmark evidence

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

Browse raw public benchmark evidence57 rows

Agentic

  • Terminal-Bench 4.1

    Claude Sonnet 512.10%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • Terminal-Bench 3.0

    Claude Sonnet 514.6%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • Terminal-Bench 2.1

    Claude Sonnet 580.4%
    Source
    Inkling 2 Mini71.6%
    Source

    Claude Sonnet 5 leads this result

  • BrowseComp

    Claude Sonnet 584.7%
    Source
    Inkling 2 Mini79%
    Source

    Claude Sonnet 5 leads this result

  • HLE w/ tools

    Claude Sonnet 557.4%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • OSWorld-Verified

    Claude Sonnet 581.2%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    Claude Sonnet 574.5%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • ApprenticeBench

    Claude Sonnet 516%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • MazeBench

    Claude Sonnet 50%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • OSWorld 2.1

    Claude Sonnet 557.0%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • Terminal-Bench 4.0

    Claude Sonnet 5—
    Inkling 2 Mini5.60%
    Source

    Not directly comparable

  • Terminal-Bench-Science 0.1

    Claude Sonnet 5—
    Inkling 2 Mini1.4%
    Source

    Not directly comparable

  • DeepSearchQA

    Claude Sonnet 5—
    Inkling 2 Mini85.3%
    Source

    Not directly comparable

  • MCP Atlas

    Claude Sonnet 5—
    Inkling 2 Mini82.7%
    Source

    Not directly comparable

  • JobBench

    Claude Sonnet 5—
    Inkling 2 Mini49.6%
    Source

    Not directly comparable

  • EnterpriseOps-Gym

    Claude Sonnet 5—
    Inkling 2 Mini42.00%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    Claude Sonnet 585.2%
    Source
    Inkling 2 Mini79.2%
    Source

    Claude Sonnet 5 leads this result

  • SWE-bench Pro

    Claude Sonnet 563.2%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • SWE Multilingual

    Claude Sonnet 578.3%
    Source
    Inkling 2 Mini73.7%
    Source

    Claude Sonnet 5 leads this result

  • SWE Multimodal

    Claude Sonnet 528.1%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • Terminal-Bench 2.1

    Claude Sonnet 580.4%
    Source
    Inkling 2 Mini71.6%
    Source

    Claude Sonnet 5 leads this result

  • FrontierCode 1.1 Main

    Claude Sonnet 542.7%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • CursorBench 3.2

    Claude Sonnet 561.5%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • VulcanBench CII v1

    Claude Sonnet 589.2%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • LiveCodeBench (Vals)

    Claude Sonnet 582.4%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • SWE-bench (Vals)

    Claude Sonnet 579.6%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • CursorBench 4.0

    Claude Sonnet 534.1%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • FrontierCode 2.0

    Claude Sonnet 5—
    Inkling 2 Mini15.1%
    Source

    Not directly comparable

  • DeepSWE

    Claude Sonnet 5—
    Inkling 2 Mini48.2%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    Claude Sonnet 5—
    Inkling 2 Mini66.9%
    Source

    Not directly comparable

  • MRCR v2 256K-512K

    Claude Sonnet 5—
    Inkling 2 Mini46.4%
    Source

    Not directly comparable

Multimodal

  • ParseBench-1

    Claude Sonnet 561.16%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • CharXiv

    Claude Sonnet 588.3%
    Source
    Inkling 2 Mini88.3%
    Source

    Tie

  • CharXiv w/o tools

    Claude Sonnet 577%
    Source
    Inkling 2 Mini80.6%
    Source

    Inkling 2 Mini leads this result

  • PerceptionBench

    Claude Sonnet 5—
    Inkling 2 Mini57.5%
    Source

    Not directly comparable

  • MMMU-Pro

    Claude Sonnet 5—
    Inkling 2 Mini74.9%
    Source

    Not directly comparable

  • OmniDocBench 1.6

    Claude Sonnet 5—
    Inkling 2 Mini92.8%
    Source

    Not directly comparable

  • BabyVision

    Claude Sonnet 5—
    Inkling 2 Mini53.4%
    Source

    Not directly comparable

  • BabyVision w/ Python

    Claude Sonnet 5—
    Inkling 2 Mini88.8%
    Source

    Not directly comparable

  • ScreenSpot Pro

    Claude Sonnet 5—
    Inkling 2 Mini76.5%
    Source

    Not directly comparable

  • Chartography (no tools)

    Claude Sonnet 5—
    Inkling 2 Mini22.1%
    Source

    Not directly comparable

  • Chartography (tools)

    Claude Sonnet 5—
    Inkling 2 Mini72.5%
    Source

    Not directly comparable

  • ZeroBench

    Claude Sonnet 5—
    Inkling 2 Mini17.0%
    Source

    Not directly comparable

  • ZeroBench w/ Python

    Claude Sonnet 5—
    Inkling 2 Mini34.0%
    Source

    Not directly comparable

  • MathVision

    Claude Sonnet 5—
    Inkling 2 Mini89.0%
    Source

    Not directly comparable

Knowledge

  • HLE

    Claude Sonnet 557.4%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • HLE w/o tools

    Claude Sonnet 543.2%
    Source
    Inkling 2 Mini30.7%
    Source

    Claude Sonnet 5 leads this result

  • HLE-Verified

    Claude Sonnet 531.0%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • LABBench2

    Claude Sonnet 580.1%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • GPQA Diamond (Vals)

    Claude Sonnet 588.9%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • MMLU-Pro (Vals)

    Claude Sonnet 587.5%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • GPQA

    Claude Sonnet 5—
    Inkling 2 Mini86.6%
    Source

    Not directly comparable

  • GPQA-D

    Claude Sonnet 5—
    Inkling 2 Mini86.6%
    Source

    Not directly comparable

Instruction following

  • MultiChallenge (provider reports)

    Claude Sonnet 5—
    Inkling 2 Mini50.00%
    Source

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    Claude Sonnet 565.614%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    Claude Sonnet 529.268%
    Source
    Inkling 2 Mini—

    Not directly comparable

  • Apex

    Claude Sonnet 5—
    Inkling 2 Mini34.4%
    Source

    Not directly comparable

57 public results · 8 shared

Watch Claude Sonnet 5 vs Inkling 2 Mini

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

Read a sample issue

Join 7,000+ subscribers.

Last updated November 21, 2026