Codestral 22B v0.1 vs Ternary Bonsai 2 27B

At a Glance

Compare
Pricing and Limits
Input priceFrom · USD / 1M tokensNot reported$0.075Openrouter · Sep 22, 2026
Output priceFrom · USD / 1M tokensNot reported$0.50Openrouter · Sep 22, 2026
Context windowMaximum documented tokens33K262K
Model facts checkedAug 28, 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

FieldCodestral-22B-v0.1Ternary Bonsai 2 27B
DeveloperMistral AIPrismML
FamilyCodestral 22b V0 1Bonsai 2
ModelCodestral-22B-v0.1Ternary Bonsai 2 27B
VersionCodestral-22B-v0.1Ternary Bonsai 2 27B
Lifecycleactiveactive
ReleasedUnknown2026-09-17
Knowledge cutoffUnknownUnknown
Input modalitiesTextText, Image
Output modalitiesTextText
Context window33K262K
Total parameters22.2B27.4B
Active parametersUnknownUnknown
Licenseotherapache-2.0
Open weightsYesYes
API availableUnknownYes
Self-hostableYesYes
Provider accessUnknownOpenrouter (Standard)
Capabilitiesgenerationchat, generation, reasoning, tools, vision
Base modelUnknownQwen3.8 27B
Effective bit widthUnknown1.76 bits per weight
Language model sizeUnknown5.93 GB
Weight formatUnknownTernary g128 with FP16 group scales

Codestral 22B v0.1 Capabilities

generation
Serving providers0
Canonical IDmistralai/Codestral-22B-v0.1

Ternary Bonsai 2 27B Capabilities

chatgenerationreasoningtoolsvision
Serving providers1
Canonical IDprism-ml/Ternary-Bonsai-2-27B

Primary Evidence

Sources and Freshness

Questions

Codestral 22B v0.1 vs Ternary Bonsai 2 27B FAQs

Is Codestral 22B v0.1 or Ternary Bonsai 2 27B better for coding?+

This comparison does not currently contain a protocol-matched coding benchmark for both Codestral 22B v0.1 and Ternary Bonsai 2 27B, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.

Which is cheaper, Codestral 22B v0.1 or Ternary Bonsai 2 27B?+

Only Ternary Bonsai 2 27B has a directly sourced input price: $0.075 per million tokens. Only Ternary Bonsai 2 27B has a directly sourced output price: $0.50 per million tokens.

Which has a larger context window, Codestral 22B v0.1 or Ternary Bonsai 2 27B?+

Ternary Bonsai 2 27B has the larger sourced context window. Codestral 22B v0.1 supports 33K and Ternary Bonsai 2 27B supports 262K.

Which performs better in benchmarks, Codestral 22B v0.1 or Ternary Bonsai 2 27B?+

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

Can Codestral 22B v0.1 or Ternary Bonsai 2 27B be self-hosted?+

Both models have the same recorded self-hosting status: supported. Codestral 22B v0.1 is open weight; Ternary Bonsai 2 27B is open weight.

Can Codestral 22B v0.1 and Ternary Bonsai 2 27B understand images?+

Codestral 22B v0.1 is not documented with image input; Ternary Bonsai 2 27B is documented with image input. This reflects supported input modalities, not vision quality.

Which can generate longer answers, Codestral 22B v0.1 or Ternary Bonsai 2 27B?+

Neither has a larger sourced maximum output. Codestral 22B v0.1 is — and Ternary Bonsai 2 27B is —.

Do Codestral 22B v0.1 and Ternary Bonsai 2 27B support reasoning and tool use?+

Codestral 22B v0.1: none of these features are definitively sourced. Ternary Bonsai 2 27B: reasoning, tool calling, and image input. Feature support does not establish relative quality.

Which is available from more inference providers, Codestral 22B v0.1 or Ternary Bonsai 2 27B?+

Codestral 22B v0.1 has 0 sourced provider routes; Ternary Bonsai 2 27B has 1, so Ternary Bonsai 2 27B has broader tracked availability.

Which offers better value, Codestral 22B v0.1 or Ternary Bonsai 2 27B?+

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