LLM Reference

Granite 3.3 8B Instruct vs Stockmark 2 100B Instruct

Granite 3.3 8B Instruct (2025) and Stockmark 2 100B Instruct (2025) are compact production models from IBM Research and Stockmark. Granite 3.3 8B Instruct ships a 128k-token context window, while Stockmark 2 100B Instruct ships a 128k-token context window. This comparison covers specs, pricing, API access, capabilities, benchmarks, input and output token costs, and production fit for coding and agent workloads.

Stockmark 2 100B Instruct is safer overall; choose Granite 3.3 8B Instruct when provider fit matters.

Decision scorecard

Local evidence first
SignalGranite 3.3 8B InstructStockmark 2 100B Instruct
Best fortool-calling agents and provider-routed productiongeneral production evaluation
Decision fitRAG, Agents, and Long contextLong context
Context window128k128k
Cheapest output$0.25/1M tokens-
Provider routes2 tracked1 tracked
Shared benchmarks0 shared0 shared

Decision tradeoffs

Choose Granite 3.3 8B Instruct when...
  • Granite 3.3 8B Instruct has broader tracked provider coverage for fallback and route flexibility.
  • Granite 3.3 8B Instruct uniquely exposes JSON / Tool use in local model data.
  • Local decision data tags Granite 3.3 8B Instruct for RAG, Agents, and Long context.
Choose Stockmark 2 100B Instruct when...
  • Local decision data tags Stockmark 2 100B Instruct for Long context.

Monthly cost at traffic

Estimate token spend from the cheapest tracked input and output route or tier on this page.

Granite 3.3 8B Instruct

$86.50

Cheapest tracked route/tier: Replicate API

Stockmark 2 100B Instruct

Unavailable

No complete token price in local provider data

Cost delta unavailable until both models have sourced input and output token prices.

Switch friction

Granite 3.3 8B Instruct -> Stockmark 2 100B Instruct
  • Provider overlap exists on NVIDIA NIM; start route-level A/B tests there.
  • Check replacement coverage for JSON / Tool use before moving production traffic.
Stockmark 2 100B Instruct -> Granite 3.3 8B Instruct
  • Provider overlap exists on NVIDIA NIM; start route-level A/B tests there.
  • Granite 3.3 8B Instruct adds JSON / Tool use in local capability data.

Specs

Specification
Released2025-03-012025-06-01
Context window128k128k
Parameters8B100B
ArchitectureDecoder OnlyDecoder Only
LicenseApache 2.0OSI-approvedOpen Weights
OpennessOpen sourceOpen weights
WeightsUnknownUnknown
CodeUnknownUnknown
Commercial useCommercial use: permitted-
Knowledge cutoff2024-04-

Pricing and availability

Pricing attributeGranite 3.3 8B InstructStockmark 2 100B Instruct
Input price$0.03/1M tokens-
Output price$0.25/1M tokens-
Providers

Capabilities

CapabilityGranite 3.3 8B InstructStockmark 2 100B Instruct
VisionNoNo
MultimodalNoNo
ReasoningNoNo
JSON / Tool useYesNo
Structured outputsNoNo
Code executionNoNo
IDE integrationNoNo
Computer useNoNo
Parallel agentsNoNo

Benchmarks

No shared benchmark scores are currently available for this pair.

Continue comparing

Last reviewed: 2026-06-29. Data sourced from public model cards and provider documentation.