# DeepSeekMath-V2 alternatives

DeepSeekMath-V2: IMO Gold level reasoning, fully open.

24 products that do a similar job, closest first. 16 have pricing checked on their own site; 8 of those are free or have a free tier.

Page: https://nowfound.app/alternatives/deepseekmath-v2
About DeepSeekMath-V2: https://nowfound.app/launch/deepseekmath-v2.md

- [DeepSeek-V3.1](https://nowfound.app/launch/deepseek-v3-1): The best-performing and best of value open-source AI model (2025-08-22, 417 votes, chat.deepseek.com)
- [DeepSeek-VL2](https://nowfound.app/launch/deepseek-vl2): MoE vision-language, now easier to access (2025-02-10, 503 votes, free, github.com)
- [DeepSeek-R1-0528](https://nowfound.app/launch/deepseek-r1-0528): New open-source LLM that rivals o3 in coding & reasoning (2025-05-29, 218 votes, from $9/mo, huggingface.co)
- [DeepSeek-V4](https://nowfound.app/launch/deepseek-v4): The open-source era of 1M context intelligence (2026-04-24, 387 votes, from $9/mo, huggingface.co)
- [DeepSeek R1](https://nowfound.app/launch/deepseek-r1): Free, Fastest, 671B Full Version, No Sign-Up (2025-01-21, 322 votes, free, github.com)
- [DeepSeek-V3.2](https://nowfound.app/launch/deepseek-v3-2): Reasoning-first models built for agents (2025-12-02, 198 votes, from $9/mo, huggingface.co)
- [DeepSeek-V3-0324](https://nowfound.app/launch/deepseek-v3-0324): Code like 3.7 but open source (2025-03-25, 273 votes, from $9/mo, huggingface.co)
- [DeepSeek-V3.2-Exp](https://nowfound.app/launch/deepseek-v3-2-exp): Long-context efficiency with DeepSeek Sparse Attention (2025-09-30, 155 votes, from $9/mo, huggingface.co)
- [OpenAI o1](https://nowfound.app/launch/openai-o1): AI that can do general-purpose complex reasoning (2024-09-13, 685 votes, openai.com)
- [Gemini 2.5 Deep Think](https://nowfound.app/launch/gemini-2-5-deep-think): Experience gold-medal AI reasoning (2025-08-02, 337 votes, blog.google)
- [OpenAI o3-mini](https://nowfound.app/launch/openai-o3-mini): Pushing the frontier of cost-effective reasoning (2025-02-01, 510 votes, openai.com)
- [Kimi K2](https://nowfound.app/launch/kimi-k2): The 1T parameter open model for agentic intelligence (2025-07-13, 345 votes, moonshotai.github.io)
- [DeepSeek-Prover-V2](https://nowfound.app/launch/deepseek-prover-v2): Advancing Formal Mathematical Reasoning via RL (2025-05-01, 7 votes, free, github.com)
- [Gemini 3 Deep Think by Google](https://nowfound.app/launch/gemini-3-deep-think-by-google): Google’s best model for logical thinking and understanding (2025-12-05, 312 votes, blog.google)
- [Llama 3.2 Interpretability with Sparse](https://nowfound.app/launch/llama-3-2-interpretability-with-sparse-autoencoders): This project replicates mechanistic interpretability research on large language models using sparse autoencoders, building on published work from Anthropic, OpenAI, and DeepMind. (2024-11-21, 579 votes, free, github.com)
- [DeepSeek-V4-Flash-0731](https://nowfound.app/launch/deepseek-v4-flash-0731): Frontier agent intelligence at Flash prices (2026-08-01, 324 votes, from $9/mo, huggingface.co)
- [OpenAI Open Models](https://nowfound.app/launch/openai-open-models): gpt-oss-120b and gpt-oss-20b open-weight language models (2025-08-06, 480 votes, openai.com)
- [TensorZero](https://nowfound.app/launch/tensorzero): TensorZero is an open-source platform for building LLM applications at scale. (2025-08-18, 307 votes, free, github.com)
- [DeepSeek-OCR](https://nowfound.app/launch/deepseek-ocr): Advanced visual-text compression technology (2025-10-21, 350 votes, from $9/mo, huggingface.co)
- [Janus](https://nowfound.app/launch/janus-st824930): Unified Multi-Modal AI by DeepSeek (2025-01-28, 224 votes, free, github.com)
- [Distilling DeepSeek into GPT-OSS doesn't](https://nowfound.app/launch/distilling-deepseek-into-gpt-oss-doesn-t-transfer-censorship): We recently used DeepSeek V4 Flash as a teacher for finance tasks with GPT-OSS-120B. Distillation works well on this problem. At a constrained 8k token budget, our self-distilled 120B scores 83.61% o… (2026-07-30, 170 votes, from $380/mo, ctgt.ai)
- [DeepSeek for iOS](https://nowfound.app/launch/deepseek-for-ios): Your Al assistant powered by DeepSeek-V3 (2025-01-15, 357 votes, free, apps.apple.com)
- [DeepSeek V3 in 200 lines of pure Jax](https://nowfound.app/launch/deepseek-v3-in-200-lines-of-pure-jax): DeepSeek V3 (MLA, DeepSeekMoE, aux-loss-free balancing, MTP) in 190 lines of pure JAX, trained on a MacBook Air. Write-up: vaibhaav.in/writing/deepseek-v3-jax (2026-11-14, 276 votes, free, github.com)
- [Qwen3.5](https://nowfound.app/launch/qwen3-5): The 397B native multimodal agent with 17B active params (2026-02-17, 288 votes, qwen.ai)
