Gemma 4, Phi-4, and Qwen3: Accuracy-Efficiency Tradeoffs in Dense and MoE Reasoning Language Models

• URL: http://arxiv.org/abs/2604.07035v1
• Date: Wed, 08 Apr 2026 12:50:52 GMT
• Title: Gemma 4, Phi-4, and Qwen3: Accuracy-Efficiency Tradeoffs in Dense and MoE Reasoning Language Models
• Abstract summary: Mixture-of-experts (MoE) language models are often expected to offer better quality-efficiency tradeoffs than dense models.<n>We present a benchmark of seven recent reasoning-oriented instruction-tuned models spanning dense and MoE designs.
• Score: 6.396911723204044
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mixture-of-experts (MoE) language models are often expected to offer better quality-efficiency tradeoffs than dense models because only a subset of parameters is activated per token, but the practical value of that advantage depends on end-to-end behavior under realistic inference constraints. We present a controlled empirical benchmark of seven recent reasoning-oriented instruction-tuned models spanning dense and MoE designs, namely Gemma-4-E2B, Gemma-4-E4B, Gemma-4-26B-A4B, Phi-4-mini-reasoning, Phi-4-reasoning, Qwen3-8B, and Qwen3-30B-A3B, evaluated on four benchmarks -- ARC-Challenge, GSM8K, Math Level 1-3, and TruthfulQA MC1 -- under three prompting strategies: zero-shot, chain-of-thought, and few-shot chain-of-thought. The study covers 8,400 total model-dataset-prompt evaluations and records accuracy, latency, peak GPU memory usage (VRAM), and an approximate floating-point operations (FLOPs)-per-token proxy. Across the weighted multi-task summary, Gemma-4-E4B with few-shot chain-of-thought achieved the best overall result, reaching weighted accuracy 0.675 with mean VRAM 14.9 GB, while Gemma-4-26B-A4B was close in accuracy at 0.663 but substantially more memory intensive at 48.1 GB. At the task level, Gemma models dominated ARC and Math, Phi models were strongest on TruthfulQA, and GSM8K showed the largest prompt sensitivity, including a sharp drop for Phi-4-reasoning from 0.67 under chain-of-thought to 0.11 under few-shot chain-of-thought. These results show that sparse activation alone does not guarantee the best practical operating point: observed accuracy-efficiency tradeoffs depend jointly on architecture, prompting protocol, and task composition. We release a reproducible benchmark pipeline, aggregated results, and paired statistical analyses to support deployment-oriented evaluation of reasoning LLMs under real resource constraints.

Related papers

• On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability [57.03393882933515]
  We describe the architecture and ablations of Qwen3.8-Flash-Next, a sparse mixture-of-experts model with 125B parameters, 6B activated per token, and 51B parameters of n-gram embedding tables held off the accelerator.<n>On fourteen pre-training benchmarks the model leads the 397B-A17B predecessor on eight and trails it on the rest by at most 2.6 points, at 1/3 the activated parameters, 1/3 the training tokens, and roughly 1/9 the training FLOPs.
  arXiv  Detail & Related papers  (2026-08-31T06:35:07Z)
• Models for minimalist RAG: B1ade 335M Embedding and 1B Parameter Small Language Models [2.1665689529884697]
  We present B1ade, an efficient RAG architecture comprising two purpose-built components: a compact embedding model and a purpose-built SLM.<n>B1ade-embed, a 335M parameter retrieval model constructed via parameter-free fusion of five pretrained encoders achieves top MTEB scores among sub-500M models with zero additional training.<n>B1ade-1B cites retrieved passages in 42.4% of responses, exceeding the attribution rate of its training distribution by 5.5 percentage points.
  arXiv  Detail & Related papers  (2026-07-29T22:41:25Z)
• Silent Failures in Quantized LLM Reasoning: A Taxonomy-Based Analysis of Hollow Convergence and Failure Mode Shifts [0.0]
  We show that post-training quantization can silently alter how large language models reason even when task accuracy is preserved.<n>We classify 30,000 chain-of-thought outputs from five instruction-tuned LLMs across three quantization precisions.<n>We find that while accuracy is robust across precisions, Hollow Convergence shows a significant size-dependent shift under NF4.
  arXiv  Detail & Related papers  (2026-07-10T21:55:05Z)
• Adapting Large Language Models to a Low-Resource Agglutinative Language: A Comparative Study of LoRA and QLoRA for Bashkir [0.0]
  This paper presents a comparative study of parameter-efficient fine-tuning methods, including LoRA and QLoRA, applied to the task of adapting large language models to the Bashkir language.<n> Experimental evaluation is conducted on a Bashkir text corpus of 71k documents (46.9M tokens) using models of various architectures.
  arXiv  Detail & Related papers  (2026-05-06T14:14:35Z)
• ACAR: Adaptive Complexity Routing for Multi-Model Ensembles with Auditable Decision Traces [3.151184728006369]
  We present ACAR, a measurement framework for studying multi-model orchestration under auditable conditions.<n>ACAR uses self-consistency variance (sigma) computed from N=3 probe samples to route tasks across single-model, two-model, and three-model execution modes.<n>We evaluate ACAR on 1,510 tasks spanning four benchmarks, producing more than 7,550 auditable runs.
  arXiv  Detail & Related papers  (2026-02-06T23:27:17Z)
• AgentCPM-Explore: Realizing Long-Horizon Deep Exploration for Edge-Scale Agents [75.67445299298949]
  AgentCPM-Explore is a compact 4B agent model with high knowledge density and strong exploration capability.<n>We introduce a holistic training framework featuring parameter-space model fusion, reward signal denoising, and contextual information refinement.<n>AgentCPM-Explore achieves state-of-the-art (SOTA) performance among 4B-class models, matches or surpasses 8B-class SOTA models on four benchmarks, and even outperforms larger-scale models such as Claude-4.5-Sonnet or DeepSeek-v3.2 in five benchmarks.
  arXiv  Detail & Related papers  (2026-02-06T08:24:59Z)
• Chain of Simulation: A Dual-Mode Reasoning Framework for Large Language Models with Dynamic Problem Routing [0.0]
  Chain of Simulation (CoS) is a novel dual-mode reasoning framework that dynamically routes problems to specialized reasoning strategies.<n>CoS employs three distinct reasoning modes: computational flow with self-consistency for mathematical problems, symbolic state tracking with representations for spatial reasoning, and hybrid fact-extraction for multi-hop inference.
  arXiv  Detail & Related papers  (2026-02-02T21:44:01Z)
• Scaling Trends for Multi-Hop Contextual Reasoning in Mid-Scale Language Models [0.0]
  We present a controlled study of multi-hop contextual reasoning in large language models.<n>We show that multi-agent systems show the inverse pattern, achieving up to 80% on reasoning tasks where rule-based methods fail.
  arXiv  Detail & Related papers  (2026-01-06T20:18:55Z)
• MobileLLM-R1: Exploring the Limits of Sub-Billion Language Model Reasoners with Open Training Recipes [60.57770396565211]
  We show that strong reasoning abilities can emerge with far less data.<n>MobileLLM-R50M achieves an AIME score of 15.5, compared to just 0.6 for OLMo-2-1.48B and 0.3 for SmolLM-2-1.7B.
  arXiv  Detail & Related papers  (2025-09-29T15:43:59Z)
• EfficientLLM: Efficiency in Large Language Models [64.3537131208038]
  Large Language Models (LLMs) have driven significant progress, yet their growing counts and context windows incur prohibitive compute, energy, and monetary costs.<n>We introduce EfficientLLM, a novel benchmark and the first comprehensive empirical study evaluating efficiency techniques for LLMs at scale.
  arXiv  Detail & Related papers  (2025-05-20T02:27:08Z)
• "Give Me BF16 or Give Me Death"? Accuracy-Performance Trade-Offs in LLM Quantization [67.3213104337679]
  Quantization is a powerful tool for accelerating large language model (LLM) inference, but the accuracy-performance trade-offs across different formats remain unclear.<n>We conduct the most comprehensive empirical study to date, evaluating FP8, INT8, and INT4 quantization across academic benchmarks and real-world tasks.
  arXiv  Detail & Related papers  (2024-11-04T18:21:59Z)
• Evaluating Large Language Models on Spatial Tasks: A Multi-Task Benchmarking Study [4.80612909282198]
  This study introduces a new multi-task spatial evaluation dataset designed to explore and compare the performance of several advanced models on spatial tasks.<n>The dataset includes twelve distinct task types, such as spatial understanding and simple route planning, each with verified and accurate answers.
  arXiv  Detail & Related papers  (2024-08-26T17:25:16Z)
• Advancing LLM Reasoning Generalists with Preference Trees [119.57169648859707]
  We introduce Eurus, a suite of large language models (LLMs) optimized for reasoning. Eurus models achieve state-of-the-art results among open-source models on a diverse set of benchmarks.
  arXiv  Detail & Related papers  (2024-04-02T16:25:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.