Model comparison · accepted catalogue facts 2026-11-22

Inkling vs Inkling Small

Neither model has both an equal-or-higher measured score and an equal-or-lower estimated price in this pair. Retaining your current model is a valid outcome.

Both models belong to the Thinking Machines Inkling collection. Inclusion does not recommend either model. This is a direct pair comparison; global frontier membership is not its verdict.

Inkling

LMArena general · Elo · higher is better

Scale starts at 1040 Elo

1443.0

Balanced effective USD / 1M tokens · lower is better

$1.76/M

Inkling Small

LMArena general · Elo · higher is better

Scale starts at 1040 Elo

1414.3

Balanced effective USD / 1M tokens · lower is better

$0.638/M

In-Depth Analysis

Inkling vs Inkling Small: Measured Technical Breakdown

Comprehensive comparison across measured human-preference evaluation, effective production token costs, architectural limits, and migration trade-offs.

Measured Capability (LMArena)
Inkling leads by 28.7 Elo

Inkling: 1443 ±3.96 · Inkling Small: 1414.3 ±4.52 Statistically significant preference separation

Cost Comparison Across Canonical Production Workloads

Rates represent effective cost per 1M tokens under standard token blends. Cached prompt reads are factored according to workload context reuse.

Workload & RatioInklingInkling SmallCost DifferenceAdvantage
Balanced 3 tokens in per 1 out$1.76 /1M tokens$0.638 /1M tokens$1.13 (64% diff)Inkling Small costs less
Summarise Huge input, tiny output$0.916 /1M tokens$0.388 /1M tokens$0.528 (58% diff)Inkling Small costs less
Chat Long cached system prompt$1.97 /1M tokens$0.645 /1M tokens$1.33 (67% diff)Inkling Small costs less
Code gen Output dominates$2.76 /1M tokens$0.872 /1M tokens$1.89 (68% diff)Inkling Small costs less
Agentic Context replayed each turn$0.964 /1M tokens$0.354 /1M tokens$0.609 (63% diff)Inkling Small costs less

Architectural & Operational Specifications

SpecificationInklingInkling SmallOperational Note
Context window524,288 tokens524,288 tokens—
Max output generation471,859 tokens262,144 tokensInkling allows longer single-turn completions
Base input price$1.00 / 1M tokens$0.450 / 1M tokensInkling Small input rate is cheaper
Base output price$4.05 / 1M tokens$1.20 / 1M tokensInkling Small output rate is cheaper
Cached input price$0.170 / 1M tokens$0.100 / 1M tokensInkling Small offers lower cached prompt cost
Reasoning & thinking tokensSupported; standard ratesSupported; standard rates—
Input modalitiestext, image, audiotext, image, audio—
Tool calling / function useSupportedSupported—
Open weights & licenceProprietary (API-only)Proprietary (API-only)—
Provider & reseller marketThinking Machines · 1 provider offersThinking Machines · 1 provider offers—

When to Choose Inkling

  • Higher measured general capability (1443 Elo on human evaluation).
  • Higher maximum completion limit (471,859 tokens).
  • Direct integration with Thinking Machines tooling and API features.

When to Choose Inkling Small

  • Direct integration with Thinking Machines tooling and API features.
Risks when moving from Inkling → Inkling Small:
  • Output limit: 471,859 → 262,144 tokens.

Compare · model decision

Choose two or three models

Inspect the differences, estimate your workload and keep the evidence. All standard catalogue models are available, including unrated ones.

Catalogue facts
2026-11-22

2 / 3 selected · 0 requirements
Open public comparison link
Public workload and requirements

These settings are included in the share link. The fixed-request scenario and private evaluation data below are excluded.

A ↔ B

Neither model has both an equal-or-higher measured score and an equal-or-lower estimated price in this pair. Retaining your current model is a valid outcome.

Estimated blend: 75% input, 25% output; 0% cached input. Base context tier. Prompt length selects the billing tier. An unlisted cache-read rate uses that tier’s full input rate in this blend; missing required input or output rates prevent an estimate. The fixed-request estimate below validates actual context and output limits and requires a published cache-read rate when cache hits are assumed.

What changes between these models

Quality first, then estimated price. LMArena scores are human-preference measurements, not a benchmark of your application. A missing score is unrated.

  • Inkling · Meets the stated model requirements
  • Inkling Small · Meets the stated model requirements
Material differences and unknown evidence. Deltas compare each model with model A. Unknown facts remain visible.
Measured factA InklingB Inkling Small
LMArena general score (Elo)1,4431,414.3Δ vs A: -28.7 Elo
Score interval half-width (± Elo)3.964.52Δ vs A: +0.56 Elo
LMArena document score (Elo)UnknownUnknown
Estimated effective USD / 1M tokens$1.7625$0.6375Δ vs A: -1.125 USD/M
Base input USD / 1M tokens$1$0.45Δ vs A: -0.55 USD/M
Base output USD / 1M tokens$4.05$1.2Δ vs A: -2.85 USD/M
Base cached-input USD / 1M tokens$0.17$0.1Δ vs A: -0.07 USD/M
Output limit (tokens)471,859262,144Δ vs A: -209,715 tokens
Open weightsUnknownUnknown
Recorded licenceUnknownUnknown
Recorded retirement dateUnknownUnknown
9 shared measured facts
Shared measured facts. Deltas compare each model with model A. Unknown facts remain visible.
Measured factA InklingB Inkling Small
Benchmark reasoning effortdefaultdefault
Context window (tokens)524,288524,288Δ vs A: 0 tokens
Input modalitiestext, image, audiotext, image, audio
Tool useSupportedSupportedΔ vs A: 0
ReasoningSupportedSupportedΔ vs A: 0
Model providerThinking MachinesThinking Machines
Distinct sellers in the catalogue11Δ vs A: 0
Serving precision differs across offersNot supportedNot supportedΔ vs A: 0
Retirement evidenceNone announced; not a guaranteeNone announced; not a guarantee

Provider counts and precision flags describe catalogue offers. They do not establish the precision, availability, region, tool behaviour or limits of your chosen endpoint. Open each model’s provider table before switching.

Complete price ladders and billing conditions

Inkling

Rates below are USD per 1M tokens. A threshold selects that whole input/output rate pair.

Input-token thresholdInputOutputCached input
Base, up to and including first threshold$1$4.05$0.17

Inkling Small

Rates below are USD per 1M tokens. A threshold selects that whole input/output rate pair.

Input-token thresholdInputOutputCached input
Base, up to and including first threshold$0.45$1.2$0.1
Migration preflight · compare each candidate with model A

Keeping Inkling is a valid decision. No pairwise comparison certifies a drop-in replacement.

Inkling → Inkling Small

Known loss or failed requirement

  • Output limit: 471,859 → 262,144 tokens.
  • Provider endpoint, serving precision, availability and task success need your own verification. Catalogue model capabilities do not certify an endpoint.
Check alternatives to Inkling

Estimate a fixed request shape

Private to this tab unless you save locally or explicitly export private details. These are estimates, not observed invoices. Cross-model token counts and task success are assumptions.

0.1 extra attempts means 10 extra full attempts per 100 requests. Cache writes, tool charges, taxes and unlisted charges are excluded.

  • InklingEstimated total: $4.025 1,000 full attempts · base rates
  • Inkling SmallEstimated total: $1.5 1,000 full attempts · base rates
Prompt and output sensitivity · sampled scenarios

Every point recalculates the entire request shape. Tier-boundary samples include one token below, at and above each threshold. Ordering between samples is not guaranteed. Lowest known cost excludes unknown estimates and is not a replacement recommendation.

Vary input; hold output, requests, cache and retries fixed.
Input tokensInklingInkling SmallCoverage
2,000$4.025$1.5Lowest knownAll computed
4,096$6.121$2.4432Lowest knownAll computed
32,768$34.793$15.3456Lowest knownAll computed
131,072$133.097$59.5824Lowest knownAll computed
Vary output; hold input, requests, cache and retries fixed.
Output tokensInklingInkling Small
128$2.5184$1.0536
2,048$10.2944$3.3576
8,192$35.1776$10.7304

Bring your observed evaluation results

JSON is parsed locally. Prompts, filenames, individual responses and extra fields are not retained. Import only aggregate measurements for selected models. Task/version labels do not establish that two evaluations used the same dataset.

Expected format
[{"model":"provider/exact-model-id","task":"support-answer","version":"rubric-1","attempts":100,"successes":84,"costUsd":2.4,"latencyMs":950}]

Use the exact catalogue ID. Replace the example values with observed measurements. Omit costUsd or latencyMs when they are unknown. No public score is inferred from these observations.

Keep the decision and its evidence

A portable record retains selected facts, complete tier ladders, public requirements and the publication manifest. It also embeds the full permitted public catalogue to verify the input artifact checksum. Files are larger for this reason (up to 2 MB); local storage is limited to 12 records. Team review is local to this browser; there is no hosted collaboration.

Saving is opt-in and includes private review fields. Shared comparison links never include them. Version 1 Check references remain separate and are not overwritten or guessed into this format.

Saved local decisions · 0

No version 2 decision records are stored in this browser.

Questions this page answers

Is Inkling a better-and-cheaper replacement for Inkling Small?

Neither model has both an equal-or-higher measured score and an equal-or-lower estimated price in this pair. Retaining your current model is a valid outcome.

Which model is higher rated: Inkling or Inkling Small?

Inkling scores higher on LMArena general (1443.0 ±4.0) compared to Inkling Small (1414.3 ±4.5), a margin of 28.7 Elo points. The measured confidence intervals do not overlap, demonstrating a statistically significant preference lead on human evaluations.

Which model is more cost-effective: Inkling or Inkling Small?

On a standard balanced workload (3 prompt tokens per 1 generated token), Inkling Small costs $0.638 per 1M tokens versus $1.76 for Inkling. Inkling is approximately 2.8× the price of Inkling Small. Base input rates are $1.00 for Inkling and $0.450 for Inkling Small; base output rates are $4.05 for Inkling and $1.20 for Inkling Small.

How do their context windows and generation limits compare?

Both Inkling and Inkling Small provide an identical 524,288-token context window. Maximum output generation limits are 471,859 tokens for Inkling and 262,144 tokens for Inkling Small.

What capabilities are lost when switching between Inkling and Inkling Small?

Switching from Inkling to Inkling Small involves: Output limit: 471,859 → 262,144 tokens. Run test suites on your specific prompt templates and tool declarations before cutting over production traffic.

Can Inkling or Inkling Small be self-hosted or deployed privately?

Both Inkling and Inkling Small are proprietary cloud models offered via managed API endpoints. Inkling is served by Thinking Machines and 1 catalogue offers, while Inkling Small is served by Thinking Machines and 1 catalogue offers.

What is included in this comparison?

Accepted model capabilities, context and output limits, complete price-tier ladders, independent LMArena measurements and missing evidence. Provider endpoint precision and application success still need verification.

Can I compare a different workload or keep a decision record?

Yes. Change the public workload, prompt length and requirements in the comparison workspace. Fixed-request scenarios, aggregate evaluations and review notes stay local unless explicitly exported.

Evidence & Ask