MiniMax M2.5
MiniMax M2.5 is a 230B MoE model (10B active) that scores 80.2% on SWE-Bench Verified while costing 1/10th to 1/20th of frontier competitors like Claude Opus 4.6 and GPT-5.2.
Articles Tagged "MoE"
Qwen 3.5 Medium Series Drops Four Models That Make the 235B Flagship Obsolete
Alibaba releases four Qwen 3.5 medium models - Flash, 35B-A3B, 122B-A10B, and 27B - that match or beat the previous 235B flagship at a fraction of the compute. The 35B model activates just 3 billion parameters and still outperforms Qwen3-235B-A22B.
Qwen3.5-122B-A10B
Qwen3.5-122B-A10B is a 122B-parameter MoE model activating 10B parameters per token, narrowing the gap between medium and frontier models with top scores in GPQA Diamond (86.6), MMMU (83.9), and OCRBench (92.1). Apache 2.0 licensed.
Qwen3.5-35B-A3B
Qwen3.5-35B-A3B is a 35B-parameter MoE model activating just 3B parameters per token that surpasses the previous Qwen3-235B flagship across language, vision, and agent benchmarks. Apache 2.0 licensed.
Qwen3.5-Flash
Qwen3.5-Flash is Alibaba's hosted production model with 1M context, built-in tools, and multimodal support at $0.10/M input tokens - one of the cheapest frontier-tier APIs available.
Liquid AI Drops LFM2-24B - A 24 Billion Parameter Model That Runs on Your Laptop
MIT spinoff Liquid AI releases LFM2-24B-A2B, a hybrid mixture-of-experts model that activates only 2.3B parameters per token, fits in 32GB RAM, and hits 112 tokens per second on a consumer CPU.
Gemini 3.1 Pro
Google DeepMind's Gemini 3.1 Pro leads on 13 of 16 benchmarks with 77.1% ARC-AGI-2, 94.3% GPQA Diamond, and a 1M-token context window at $2/M input.
Alibaba Drops Qwen3.5: 397B Parameters, 17B Active, and It Trades Blows with GPT-5.2
Alibaba's Qwen team releases Qwen3.5-397B-A17B, an open-weight mixture-of-experts model with native multimodal support and a hybrid attention architecture that runs 8x faster than its predecessor. Apache 2.0 licensed.
StepFun's Step 3.5 Flash Uses Just 11B of Its 196B Parameters and Still Rivals GPT-5.2
Shanghai AI lab StepFun open-sources Step 3.5 Flash, a 196B sparse MoE model that activates only 11B parameters per token while matching frontier models on reasoning, coding, and agentic benchmarks.
GPT-OSS 20B
OpenAI's open-weight 21B MoE reasoning model with 131K context, Apache 2.0 license, and o3-mini-level benchmark performance running in 16 GB of memory.
