Granite 4.2 8B vs Ternary Bonsai 1.7B

At a Glance

Compare
Pricing and Limits
Input priceFrom · USD / 1M tokens$0.060Deepinfra · Sep 23, 2026Not reported
Output priceFrom · USD / 1M tokens$0.25Deepinfra · Sep 23, 2026Not reported
Context windowMaximum documented tokens131K33K
Model facts checkedSep 2, 2026View model evidence →Sep 18, 2026View model evidence →

Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →

Available Benchmarks

All benchmark results →
No Protocol-Matched Benchmark Yet.Results appear here only when both models share the same benchmark version, metric, evaluation protocol, and evidence class.

Side-by-Side Facts

FieldGranite 4.2 8BTernary Bonsai 1.7B
DeveloperIBMPrismML
FamilyGranite 4 2Bonsai 1 7b
ModelGranite 4.2 8BTernary Bonsai 1.7B
VersionGranite 4.2 8BTernary Bonsai 1.7B
Lifecycleactiveactive
Released2026-08-252026-04-18
Knowledge cutoffUnknownUnknown
Input modalitiesTextText
Output modalitiesTextText
Context window131K33K
Total parameters8.8B1.7B
Active parametersUnknownUnknown
Licenseapache-2.0apache-2.0
Open weightsYesYes
API availableYesNo
Self-hostableYesYes
Provider accessDeepinfra (Standard), Openrouter (Standard)Unknown
Capabilitieschat, generation, reasoning, structured_outputs, toolschat, generation
Effective bit widthUnknown1.58 bits per weight
Weight sizeUnknown0.46 GB
Weight formatUnknownTernary Q2_0

Granite 4.2 8B Capabilities

chatgenerationreasoningstructured outputstools
Serving providers2
Canonical IDibm-granite/granite-4.2-8b

Ternary Bonsai 1.7B Capabilities

chatgeneration
Serving providers0
Canonical IDprism-ml/Ternary-Bonsai-1.7B

Primary Evidence

Sources and Freshness

Questions

Granite 4.2 8B vs Ternary Bonsai 1.7B FAQs

Is Granite 4.2 8B or Ternary Bonsai 1.7B better for coding?+

This comparison does not currently contain a protocol-matched coding benchmark for both Granite 4.2 8B and Ternary Bonsai 1.7B, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.

Which is cheaper, Granite 4.2 8B or Ternary Bonsai 1.7B?+

Only Granite 4.2 8B has a directly sourced input price: $0.060 per million tokens. Only Granite 4.2 8B has a directly sourced output price: $0.25 per million tokens.

Which has a larger context window, Granite 4.2 8B or Ternary Bonsai 1.7B?+

Granite 4.2 8B has the larger sourced context window. Granite 4.2 8B supports 131K and Ternary Bonsai 1.7B supports 33K.

Which performs better in benchmarks, Granite 4.2 8B or Ternary Bonsai 1.7B?+

There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.

Can Granite 4.2 8B or Ternary Bonsai 1.7B be self-hosted?+

Both models have the same recorded self-hosting status: supported. Granite 4.2 8B is open weight; Ternary Bonsai 1.7B is open weight.

Can Granite 4.2 8B and Ternary Bonsai 1.7B understand images?+

Granite 4.2 8B is not documented with image input; Ternary Bonsai 1.7B is not documented with image input. This reflects supported input modalities, not vision quality.

Which can generate longer answers, Granite 4.2 8B or Ternary Bonsai 1.7B?+

Neither has a larger sourced maximum output. Granite 4.2 8B is — and Ternary Bonsai 1.7B is —.

Do Granite 4.2 8B and Ternary Bonsai 1.7B support reasoning and tool use?+

Granite 4.2 8B: reasoning and tool calling. Ternary Bonsai 1.7B: none of these features are definitively sourced. Feature support does not establish relative quality.

Which is available from more inference providers, Granite 4.2 8B or Ternary Bonsai 1.7B?+

Granite 4.2 8B has 2 sourced provider routes; Ternary Bonsai 1.7B has 0, so Granite 4.2 8B has broader tracked availability.

Which offers better value, Granite 4.2 8B or Ternary Bonsai 1.7B?+

There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.

Send Feedback