The frontier of open-weight model releases
Open-weight model releases tracked by RunLocalAI — recent additions, rising families, distill chains, multimodal and reasoning waves. Each card links into the catalog with authority badges (L1.25 enriched · benchmark-backed · verdict) so you can scan editorial coverage at a glance.
Filtered results (21)
Models matching your filters. Clear filters by clicking “Any” on each row above, or remove individual filters via the URL.
Mistral Medium 3 24B (dense)
research / non-commercial workstation deployments
Mistral Small 3.2 24B
consumer-tier multilingual instruction-following
Magistral 32B
research / non-commercial reasoning at 32B scale
Devstral Small 2 24B
Apache 2.0 coding alternative to Qwen 2.5 Coder
Mistral Saba 24B
Arabic / South-Asian multilingual
Mistral Small 3 24B
consumer-tier multilingual instruction-following — Mistral's instruction-tuned baseline at 24B
Ministral 8B Instruct
consumer-tier long-context — research only
Codestral Mamba 7B
long-context coding workloads where memory matters
Codestral 22B
workstation coding at 22B class
Sarvam M
Hindi and Indian-language reasoning or chat tasks on capable local hardware
Bielik 11B v2.3 Instruct
Polish-language instruction following and chat
Bielik 11B v2.2 Instruct GGUF
Polish-language instruction following and document Q&A
Bielik 11B v2.3 Instruct
Polish-language instruction following and text generation
Japanese StableLM Instruct Gamma 7B
Japanese instruction-following and document chat
Turkish Mistral 7B Instruct v0.2
Mistral 7B Instruct v0.2
General-purpose instruction following in commercial apps on a tight VRAM budget
Mistral 7B Instruct v0.1
Lightweight local instruction-following experiments
Mistral 7B OpenOrca GGUF
English instruction-following on CPU or low-VRAM hardware
Bielik 7B Instruct v0.1 GGUF
Personal or research Polish-language chat and instruction tasks
Mistral 7B Instruct v0.2
Local instruction-following and document Q&A with long context
Bielik 7B v0.1
Polish-language fine-tuning base for downstream tasks
Going deeper
- Ecosystem maps — structured-landscape views (memory frameworks, inference runtimes, MCP, coding agents).
- Execution stacks — recipes that combine models with runtimes + hardware.
- Frontier index — broader ecosystem-momentum view across coding agents, inference runtimes, memory systems, MCP.
- Benchmarks — measured tokens-per-second + topology fields across hardware/model/runtime triples.