Phi-4 Mini Instruct vs Kimi K2.7 Code

At a Glance

Compare
Kimi K2.7 CodeMoonshot AI
Intelligence, Cost, and Efficiency
CostLower is better · Published-token output estimateUnrankedNot in the 44-model eligible cohort#9 of 44$0.042 per LiveBench case
Pricing and Limits
Input priceFrom · USD / 1M tokensNot reported$0.68Deepinfra · Sep 21, 2026
Output priceFrom · USD / 1M tokensNot reported$3.21Openrouter · Sep 22, 2026
Context windowMaximum documented tokens131K262K
Model facts checkedAug 28, 2026View model evidence →Sep 3, 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

FieldPhi-4-mini-instructKimi K2.7 Code
DeveloperMicrosoftMoonshot AI
FamilyPhi 4 Mini InstructKimi K2 7
ModelPhi-4-mini-instructKimi K2.7 Code
VersionPhi-4-mini-instructKimi K2.7 Code
Lifecycleactiveactive
Released2025-02-262026-06-11
Knowledge cutoffUnknownUnknown
Input modalitiesTextText, Image, Video
Output modalitiesTextText
Context window131K262K
Total parameters3.8B1T
Active parametersUnknown32B
Licensemitmodified-mit
Open weightsYesYes
API availableYesYes
Self-hostableYesYes
Provider accessHugging Face (Standard)Deepinfra (Standard), Fireworks Ai (Standard), Openrouter (Standard), Together Ai (Standard)
Capabilitieschat, generationagents, chat, coding, reasoning, tools, vision

Phi-4 Mini Instruct Capabilities

chatgeneration
Serving providers1
Canonical IDmicrosoft/Phi-4-mini-instruct

Kimi K2.7 Code Capabilities

agentschatcodingreasoningtoolsvision
Serving providers4
Canonical IDmoonshotai/Kimi-K2.7-Code

Primary Evidence

Sources and Freshness

Questions

Phi-4 Mini Instruct vs Kimi K2.7 Code FAQs

Is Phi-4 Mini Instruct or Kimi K2.7 Code better for coding?+

This comparison does not currently contain a protocol-matched coding benchmark for both Phi-4 Mini Instruct and Kimi K2.7 Code, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.

Which is cheaper, Phi-4 Mini Instruct or Kimi K2.7 Code?+

Only Kimi K2.7 Code has a directly sourced input price: $0.68 per million tokens. Only Kimi K2.7 Code has a directly sourced output price: $3.21 per million tokens.

Which has a larger context window, Phi-4 Mini Instruct or Kimi K2.7 Code?+

Kimi K2.7 Code has the larger sourced context window. Phi-4 Mini Instruct supports 131K and Kimi K2.7 Code supports 262K.

Which performs better in benchmarks, Phi-4 Mini Instruct or Kimi K2.7 Code?+

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

Can Phi-4 Mini Instruct or Kimi K2.7 Code be self-hosted?+

Both models have the same recorded self-hosting status: supported. Phi-4 Mini Instruct is open weight; Kimi K2.7 Code is open weight.

Can Phi-4 Mini Instruct and Kimi K2.7 Code understand images?+

Phi-4 Mini Instruct is not documented with image input; Kimi K2.7 Code is documented with image input. This reflects supported input modalities, not vision quality.

Which can generate longer answers, Phi-4 Mini Instruct or Kimi K2.7 Code?+

Neither has a larger sourced maximum output. Phi-4 Mini Instruct is — and Kimi K2.7 Code is —.

Do Phi-4 Mini Instruct and Kimi K2.7 Code support reasoning and tool use?+

Phi-4 Mini Instruct: none of these features are definitively sourced. Kimi K2.7 Code: reasoning, tool calling, and image input. Feature support does not establish relative quality.

Which is available from more inference providers, Phi-4 Mini Instruct or Kimi K2.7 Code?+

Phi-4 Mini Instruct has 1 sourced provider route; Kimi K2.7 Code has 4, so Kimi K2.7 Code has broader tracked availability.

Which offers better value, Phi-4 Mini Instruct or Kimi K2.7 Code?+

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