All Open Weights tools (60)
- ApodexSelf-evolving heavy-duty solver for deep research and agentic work. Splits reasoner from verifier, runs a trained Agent Team, and ships open weights plus the FrontierAgent CLI research harness.
- Agnes-3.0-FlashAgnes-3.0-Flash Preview is an Apache-2.0 33B multimodal open-weight checkpoint with 262k context, not the proprietary API model.
- LensVLM-9BApple's 9B vision-language model reads documents as compressed images, then expands only the pages that matter, holding accuracy to 4.3x compression.
- Bespoke NimbleBespoke Labs' open-weight 9B model and training recipe that turn Jev-style decisions into single-token choices with probabilities, runnable locally.
- BAAI bge-reranker-v2.5-gemma2-lightweightPrevious BGE reranker v2.5 Gemma2 lightweight snapshot.
- Breeze TTS 2Open-weight multilingual TTS from BreezeBlue, built for real-time interaction. Reference-free voice design and reference-guided voice direction, with sub-40m.
- cogvlm-base-490-hfPrevious CogVLM snapshot. Not a 2026 VLM default.
- Cohere Command A+Command A+ is Cohere's first mixture-of-experts flagship: 218B total and 25B active parameters, 48 languages, vision input, Apache 2.0 open weights.
- DeepSeek V4 Pro 0813DeepSeek flagship 1.6T MoE with 49B active, 1M context, MIT weights. API $0.29/$1.20 cached $0.03.
- DFlash 2Inco AI's Apache 2.0 block-diffusion drafter for speculative decoding. Released 2026-08-18 for Qwen3.8-27B and Muse Glimmer.
- Gemma 4 26B A4BGoogle DeepMind's open-weight 25.2B MoE model with 3.8B active parameters, 256K context, multimodal input, and a commercially permissive Apache 2.0 license.
- GLM-5.3-FlashZhipu's first natively multimodal GLM (confirmed as the 'Ox Alpha' stealth model): 320B MoE with 18B active, 1M context, MIT open weights at $0.15/M input.
- Google: Gemma 2 27BPrevious Gemma 2 27B snapshot. Not a 2026 default open model.
- Google: Gemma 2 9BPrevious Gemma 2 9B snapshot. Not a 2026 default small model.
- IBM Granite 4.2 30BOpen-weight dense reasoning LLM from IBM: 30B parameters, Apache 2.0, native 128K context extendable to 512K, built-in chain-of-thought and reasoning-augmented tool calling.
- Hy4 previewTencent's next-generation open-weight MoE flagship: 770B total params, 49B activated per token, 1M-token context. Apache 2.0, Gated DSA attention, and blind-eval scores edging out GLM 5.3 and Kimi K3.
- KAT-Coder V2.5Kwaipilot's open-weight agentic coding model: 35B MoE, 3B active per token, Qwen3.6 base, Apache 2.0, 262K context, top PinchBench tool-use score.
- Kimi K3Moonshot AI's open-weight 2.8T multimodal agentic model with 1M-token context, the world's first open 3T-class model rivaling closed frontier models.
- Laguna S 2.1Poolside's open-weight 118B MoE coding model with 8B active parameters, a 1M-token context window, native interleaved reasoning, and an OpenMDW-1.1 license.
- LFM 2.5-2.6BLiquid AI's open-weight 2.6B dense model trained for on-device agentic workloads, with 128K context and native tool calling.
- Ling-3.0-flash-VLLing-3.0-flash-VL is inclusionAI's MIT multimodal MoE: 124B total, 5.5B active, with image and video.
- Ling-3.0-flashinclusionAI MIT hybrid-linear MoE: 124B total, 5.1B active.
- Ling-3.0-tinyLing-3.0-tiny is inclusionAI's MIT hybrid-linear MoE: 7.9B total, 1.3B active, for local agents.
- Llama 3.1 Euryale 70B v2.2Previous Llama 3.1 Euryale 70B v2.2 snapshot. Not a 2026 Llama default.
- Llama 3 Euryale 70B v2.1Previous Llama 3 Euryale 70B v2.1 snapshot. Not a 2026 Llama default.
- llava-v1.6-34b-hfPrevious LLaVA-NeXT / LLaVA 1.6 snapshot.
- LongCat 2.0Meituan's open-weight MoE model: 1.6T total / ~48B active params, 1M context, MIT. Strong on coding and agentic tasks (2026-08-25).
- Llama v2 70B ChatPrevious Llama v2 70B Chat snapshot. Not a 2026 chat default.
- MiMo-V2.6Xiaomi's MiMo-V2.6 open-weights series pairs a 1.02T-parameter omni MoE with a 309B Flash tier, both MIT licensed with 1M-token context.
- MiniCPM5-2BOpenBMB MiniCPM5-2B is a dense 2B on-device LLM with 131k context, Apache-2.0 weights, and cookbooks for vLLM, SGLang, and llama.cpp.
- MiniMax H3MiniMax open multimodal video model (July 31, 2026): text/image/video/audio in, 768P or 2K out, 4-15s clips, native stereo, pay-as-you-go API.
- MiniMax M3MiniMax open-weight frontier model (June 1, 2026): 1M-token MSA context, native image/video input, desktop computer use, and discounted API pricing.
- MiniMax-Music3MiniMax's open-weight music generation model with studio-grade sound, natural vocal synthesis, and full-song generation from prompts and lyrics, with official ComfyUI support.
- Mistral: Mistral NemoPrevious Mistral Nemo 12B snapshot. Live catalog is Medium 3.5 / Small 4 / Large 3.
- Mistral Nemo Inferor 12BPrevious Inferor 12B snapshot. Live catalog is Medium 3.5 / Small 4 / Large 3.
- MythoMax 13BPrevious MythoMax 13B snapshot.
- Nex-N2.5-miniNex-N2.5-mini is Nex-AGI's Apache-2.0 multimodal agent model for computer use, browsing, and coding.
- Nex-N2.5-ProNex-N2.5-Pro is Nex-AGI's Apache-2.0 multimodal agent checkpoint for computer use, browsing, and coding.
- Nous: Hermes 3 405B InstructPrevious Hermes 3 405B Instruct snapshot. Not a 2026 frontier default.
- NousResearch: Hermes 2 Pro - Llama-3 8BPrevious Hermes 2 Pro Llama-3 8B snapshot.
- NVIDIA: Llama 3.1 Nemotron 70B InstructPrevious NVIDIA Llama 3.1 Nemotron 70B Instruct snapshot. Not a 2026 default.
- NVIDIA Nemotron 3.5 Lightning 30B A3BNVIDIA's efficient open-weight 30B MoE hybrid model with 3B active parameters, 1M-token context, and single-GPU deployment for local reasoning and coding.
- Qwen-Drive-1.0-4BQwen-Drive-1.0-4B is Alibaba's Apache-2.0 driving VLM that unifies 3D perception, VQA, and motion planning.
- Qwen-Image-2.1Qwen-Image-2.1 is Alibaba's 7B open-weights model that both generates and edits images, including native transparent PNGs and edits from up to ten references.
- Qwen3.8-2.4T-A95BAlibaba Qwen3.8 2.4T MoE with 95B active, 1M context, multimodal input. Coder $0.14/$0.80; 397B $0.29/$1.20.
- Qwen3.8-27BAlibaba's open-weight 27B dense companion to Qwen3.8-Max: Apache 2.0 license, multimodal image-text input, built for local deployment and small-batch inference.
- Qwen3.8-Flash-NextAlibaba's open-weight architecture preview of Qwen4: 125B multimodal MoE with 6B active plus a 51B N-gram table, 262K native context, at $0.16/M input.
- Qwen3.8-MaxAlibaba's 2.4T-parameter open-weight flagship: 95B active MoE, text/image/video input, 1M context, at $2/$6 per million tokens.
- Realtime-VenusRealtime-Venus is an Apache 2.0 9B full-duplex audio-visual model that can speak, interrupt, and delegate while it still listens.
- Rocinante 12BPrevious Rocinante 12B snapshot.
- sanoTTSTiny neural TTS from 294k to 2.3M parameters. Runs in the browser or on a $3 ESP32-S3. 11 voices, 6 languages, GPL-3.0.
- Spark-X2.5-4BXHToken's efficient 4B open-weight model: hybrid sliding-window attention for native 1M-token context, strong coding and agent performance, Apache 2.0, multilingual.
- Supra2-IMGSupra2-IMG is an Apache-2.0, 104M-parameter text-to-image DiT trained from scratch in nine hours on one H100, producing 256x256 images.
- Swift-Qwen3.8-27BSwift-Qwen3.8-27B is UkisAI's reasoning-efficient Qwen3.8-27B finetune that cuts thinking tokens about 58% with under 1% score loss.
- Ternary Bonsai 2Ternary Bonsai 2 is a 27B Qwen3.8 ternary GGUF that the model card says keeps 98.2% of FP16 scores in a 5.95 GB pack.
- Tiel-Coder 35B A3BAgentic coding MoE. Not a lab release.
- VonVon is an Apache-2.0 395M System One decision model that answers typed questions locally in about 18 ms on Apple MPS, as an open answer to TypeSafe Jev.
- WizardLM-2 7BPrevious WizardLM-2 7B snapshot. Do not keep leftover 10x-larger crown.
- WizardLM-2 8x22BPrevious WizardLM-2 8x22B snapshot.
- YuE2-3BYuE2-3B is M·A·P's CC BY-NC music model that plans an editable score, then renders vocals and accompaniment locally.
All Tags
A2A ProtocolAgent FrameworkAgent OrchestrationAgent RuntimeAgent WalletAI AgentAI AssistantAI ContentAI PaymentsAI SearchAlibabaAnthropicAPIAutomationAWSBAAIBlack Forest LabsBrowser AutomationByteDanceChatbotClaudeClaude CodeCLICloudCloudflareCode AssistantCode QualityCode ReviewCodexCoding AgentCohereCommunicationComputer UseCrypto & Web3CursorData ScienceData VisualizationDatabaseDebuggingDeep ResearchDeepgramDeepSeekDeploymentDesignDeveloper ToolsDocument ProcessingDocumentationEducationElevenLabsEmbeddingEnterpriseEvaluationFine-TuningFLUXFrontier ModelGeminiGemmaGitGoGoogleGPTGPUGrokHealthHugging FaceIBMImage GenerationIntegrationJina AIKimiKnowledge BaseLangChainLangGraphLinuxLlamaLocal AILong ContextmacOSMarkdownMCPMetaMicrosoftMiniMaxMistralMixedbread AIMixture of ExpertsModel RoutingMoonshot AIMulti-AgentMultilingualMultimodalNo-Code / Low-CodeNous ResearchNVIDIAObservabilityObsidianOpen SourceOpen Weights
Items tagged with Open Weights (60)
Month Visit45000000DR: 95AS: 88
Month Visit20000000DR: 90AS: 95
Month Visit20000000DR: 90AS: 95
Month Visit2500000DR: 70AS: 60
Month Visit1000000DR: 85AS: 85
Month Visit500000DR: 70AS: 72
Month Visit345DR: 100AS: 100
Month Visit345DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit100DR: 100AS: 100
Month Visit215DR: 92AS: 87
Month Visit100DR: 90AS: 85