Stockmark
1 model across 1 family · Latest: Stockmark 2 100B Instruct (2025-06)
AI-Powered Business Insights
Stockmark's portfolio covers 1 active model across 1 current family, spanning long context. Open a model detail page to compare provider routes and sourced benchmarks.
Covers 1 workload area across 1 active tracked model; last verified 2026-06-29.
Use it for
- Teams evaluating long context across this lab's releases
- Comparing model families before committing to a flagship
- Migration and pricing follow-ups across 1 tracked models
Do not use it for
- Choosing a hosting provider without opening a model page for price ladders
Active models
1
Current models from this lab, excluding deprecated ones
Active families
1
Current model families from this lab
Open catalog
1 open
0 open source / 1 open weights
Lowest output price
Not tracked
No provider output pricing linked yet
Latest dated release
2025-06-01
Stockmark 2 100B Instruct
Freshness
2026-06-29
Researched 19d ago
Information
Release cadence
Showing 1 recent dated release (full timeline below). Latest: Stockmark 2 100B Instruct (2025-06-01).
Where this lab wins
- Long-context: 1 tracked model with context-token or InfiniteBench-class signal.
Flagship quality / price signal
Flagship: Stockmark 2 100B Instruct (best sourced coding quality-per-dollar in this portfolio).
Quality-per-dollar unavailable for this flagship — benchmark coverage or output token pricing is still missing.
Stockmark is a Japanese AI research organization founded in 2016. AI-Powered Business Insights. Stockmark ships 1 model family totaling 1 model, with the most recent release Stockmark 2 100B Instruct in 2025-06. Notable families include Stockmark. Use it as a stable reference for lab background, release coverage, and follow-up model pages as they are added. Researchers and evaluators can scan counts, links. View official API endpoints, benchmark performance, and coding/agent fit for every Stockmark model.
About
Stockmark, established in 2016 and based in Tokyo, Japan, is an AI-driven company focused on enabling data-driven decisions for businesses, particularly by extracting insights from qualitative data like reports and meeting minutes. Their primary offerings, 'Anews' and 'Astrategy,' leverage AI-powered text analysis to process vast amounts of information, with 'Anews' filtering articles from 30,000 media sources to deliver relevant content. 'Astrategy' assists companies in analyzing market trends and competitor actions by structuring information from keyword searches. The company's innovative approach centers on addressing the challenges of utilizing qualitative data, especially within the Japanese business context. Stockmark has trained its LLMs on approximately 220 billion tokens of Japanese market and management data, creating an AI system uniquely suited to understanding the nuances of the Japanese business environment. This expertise allows for more accurate insights and tailored solutions compared to generic LLMs. Stockmark's services directly tackle the bottleneck of information gathering and document creation, which constitutes a significant portion of white-collar work in Japan. By transforming unstructured qualitative data into actionable insights, Stockmark helps companies accelerate their decision-making processes and improve operational efficiency.
Featured models
| Model | Released | Context | Input price ($/1M) | Output price ($/1M) | License | Openness |
|---|---|---|---|---|---|---|
| Stockmark 2 100B Instruct | 2025-06-01 | 128k | - | - | Open Weights | Open weights |
Model families
Recent releases
- Stockmark 2 100B Instruct- 2025-06-01
Top comparisons
- ELYZA Japanese Llama 2 13B vs Stockmark 2 100B Instruct58
- ELYZA Japanese Llama 2 7B vs Stockmark 2 100B Instruct52
- Hunyuan Hy3 Preview vs Stockmark 2 100B Instruct51
- Xiaomi MiMo-V2.5 vs Stockmark 2 100B Instruct45
- Together AI - Gemma 3n-e4B vs Stockmark 2 100B Instruct38
- Llama 2 7B vs Stockmark 2 100B Instruct20
- Together AI - Llama 3 8B Lite vs Stockmark 2 100B Instruct11
- GPT-1 vs Stockmark 2 100B Instruct11
FAQ
Who founded Stockmark and when?
Stockmark was founded in 2016 and is associated with Tokyo, Japan.
What models has Stockmark released?
Stockmark ships 1 model across 1 family: Stockmark.
Is Stockmark's technology open source?
All tracked models are released under Open Weights.
Where is Stockmark headquartered?
Stockmark is headquartered in Tokyo, Japan.
What is Stockmark known for?
AI-Powered Business Insights. Its most prominent tracked family is Stockmark.
How can I access Stockmark's models?
Stockmark's models are available via NVIDIA NIM.
Explore related pages
Last reviewed: 2026-06-29. Data sourced from public lab announcements and provider documentation.