rockmatrix-ai/EdgeRock

Maintainer? Show your league in your README

Card: your league's emblem, from Copper to Legendary

[![bestofgithub](https://bestofgithub.dev/badge/rockmatrix-ai/EdgeRock.svg?style=card)](https://bestofgithub.dev/repo/rockmatrix-ai/EdgeRock)

Flat: fits beside your other badges

[![bestofgithub](https://bestofgithub.dev/badge/rockmatrix-ai/EdgeRock.svg)](https://bestofgithub.dev/repo/rockmatrix-ai/EdgeRock)

Loading README…

Model repository	Format / runtime
EdgeRock-4B (original)	BF16 · text and images
EdgeRock-4B-NF4	NF4 · included compact loader
EdgeRock-4B-INT8	INT8 · included compact loader
EdgeRock-4B-FP8	FP8 · included compact loader
EdgeRock-4B-AWQ	AWQ 4-bit · included packed INT4 GPU runtime
EdgeRock-4B-GPTQ	GPTQ 4-bit · included packed INT4 GPU runtime
EdgeRock-4B-NVFP4	NVFP4 W4A16 · BF16 reference loading

Model repository	Format / runtime
EdgeRock-4B-MLX-8bit	8-bit · group 64
EdgeRock-4B-MLX-6bit-g64	6-bit · group 64
EdgeRock-4B-MLX-4bit-g32	4-bit · group 32
EdgeRock-4B-MLX-4bit-mixed	Mixed 4-bit

Model repository	Format / runtime
EdgeRock-4B-GGUF	7 profiles · custom EdgeRock backend
EdgeRock-4B-ONNX-Q4-g32	4-bit · ONNX Runtime CUDA
EdgeRock-4B-ONNX-Q4-CPU-g32	4-bit · ONNX Runtime CPU
EdgeRock-4B-ONNX-Q8	8-bit · ONNX Runtime CUDA

Model repository	Format / runtime
EdgeRock-4B-LiteRT-LM-INT8	INT8 · LiteRT-LM CPU preview
EdgeRock-4B-LiteRT-LM-INT4	INT4 · LiteRT-LM CPU preview

Model	29-benchmark mean	AIME mean	MATH-500	MMLU-Pro	LiveCodeBench v6	IFEval
EdgeRock-4B	77.2	82.9	97.5	79.8	61.6	88.0
AHN-GDN Qwen2.5 14B	53.4	11.4	80.0	63.5	32.9	77.4
RWKV-7 G1j 13.3B	52.3	18.8	77.2	47.6	30.7	77.8
Phi-3 Medium 4K Instruct (W2047)	45.4	2.2	49.0	54.0	19.1	61.4
RWKV-7 G1j 7.2B	43.5	8.1	64.2	46.2	21.3	66.5
RWKV-7 G1i 7.2B	42.9	9.4	68.6	44.9	18.8	53.4
AHN-GDN Qwen2.5 3B	40.9	4.2	66.2	42.8	14.6	58.6
RWKV-7 G1j 2.9B	37.0	2.2	48.2	37.9	12.1	56.4
RWKV-7 G1i 2.9B	35.3	1.8	43.8	37.8	11.6	43.1
QRWKV7 7B Instruct	32.6	0.0	41.0	23.0	12.4	59.0
Falcon3-Mamba 7B Instruct	31.6	1.2	41.8	26.1	9.3	64.7
RecurrentGemma 9B IT	29.9	0.4	22.8	13.4	7.9	39.7
RWKV-7 G1j 1.5B	28.4	1.1	31.4	16.6	6.3	46.2
RWKV-7 G1i 1.5B	27.9	0.3	28.2	13.3	7.2	37.0
ARWKV-R1 7B	18.1	0.1	33.2	8.8	2.6	10.4
xLSTM 7B	16.0	0.0	25.0	4.6	2.2	34.0
ARWKV-R1 1.5B	9.9	0.0	13.6	7.5	0.3	11.1

Model	AIME mean	MMLU-Pro	SuperGPQA	GPQA-Diamond	LiveCodeBench v6	IFEval
EdgeRock-4B	82.9	79.8	60.0	64.6	61.6	88.0
Qwen3.5-4B	86.1	77.7	52.8	77.3	75.7	89.6
Gemma4-E4B-it	40.1	69.0	41.1	55.6	69.1	84.3
Nemotron-3-Nano-4B	57.1	56.8	37.4	43.4	66.4	89.5
Phi-4-mini-reasoning	38.9	38.5	32.2	48.0	35.3	45.7
MiniCPM5-2B	87.8	65.3	40.9	67.7	83.3	86.3
LFM2.5-2.6B	56.2	59.9	27.1	58.2	59.7	93.3

Benchmark	EdgeRock-4B	Qwen3.5-4B	Retained (%)
AIME 2024	86.7	89.6	96.8
AIME 2025	77.4	80.8	95.8
AIME 2026	84.6	87.9	96.2
HMMT Feb 2025	71.8	74.6	96.2
HMMT Nov 2025	69.8	78.3	89.1
HMMT Feb 2026	63.6	65.9	96.5
MATH-500	97.5	98.0	99.5
AIME three-year mean	82.9	86.1	96.3

Benchmark	EdgeRock-4B	Qwen3.5-4B
GSM8K	94.5	94.8
MATH-500	97.5	98.0
AIME 2024	86.7	89.6
AIME 2025	77.4	80.8
AIME 2026	84.6	87.9
HMMT Feb 2026	63.6	65.9
MMLU	88.2	85.2
MMLU-Pro	79.8	77.7
MMLU-Redux	92.0	89.8
CMMLU	73.3	83.9
C-Eval	74.6	84.0
GPQA-Diamond	64.6	77.3
SuperGPQA	60.0	52.8
WinoGrande	77.9	87.1
HellaSwag	64.9	80.6
ARC-C	95.5	96.8
ARC-E	98.0	98.4
BBH	85.4	86.2
BBEH	28.9	34.6
HumanEval	96.7	86.0
MBPP	87.8	71.0
LiveCodeBench v6	61.6	75.7
IFEval	88.0	89.6
IFBench	65.3	60.3
DROP	77.2	76.0
MuSR	60.4	64.9
LogiQA2	72.2	82.2
HMMT Feb 2025	71.8	74.6
HMMT Nov 2025	69.8	78.3

Benchmark	EdgeRock-4B	Qwen3.5-4B	Retained (%)
MMBench EN	88.3	90.1	98.0
MMBench CN	89.0	91.4	97.4
POPE	85.4	88.8	96.2
MME	83.0	86.8	95.6
ChartQA	83.9	90.3	92.9
AI2D	80.0	87.0	92.0
Equal-weight mean	84.9	89.1	95.3

MME metric	EdgeRock-4B	Qwen3.5-4B
Perception (points)	1,600.4	1,738.3
Cognition (points)	728.2	751.8
Total (points)	2,328.7	2,490.1

Model	32K context (MiB)	64K context (MiB)	128K context (MiB)	256K context (MiB)	History growth
EdgeRock-4B	72	72	72	72	O(1)
Qwen3.5-4B	1,024	2,048	4,096	8,192	O(N)

Component	MiB per sequence
Committed RMA anchors	32.00
Pending RMA candidates	8.00
Local-window KV	31.97
FP16 recurrent matrices	24.00
Convolution state	1.50
Metadata	0.29
Total persistent history	97.76

Model	4K input	8K input	16K input	32K input
EdgeRock-4B	12,479 (512)	12,325 (512)	12,330 (512)	12,124 (512)
Qwen3.5-4B	3,103 (64)	2,853 (64)	2,472 (64)	1,889 (64)
Gemma 4 E4B	3,940 (128)	3,432 (128)	2,880 (128)	2,196 (64)
MiniCPM5-2B	3,152 (64)	2,938 (64)	2,575 (32)	2,117 (32)
LFM2.5-2.6B	1,673 (64)	1,771 (64)	1,695 (32)	1,727 (64)

Model	Decode tokens/s
EdgeRock-4B	233.7
Qwen3.5-4B	151.3
Gemma 4 E4B	133.8
MiniCPM5-2B	164.5
LFM2.5-2.6B	229.6

Batch size	Weighted decode tokens/s	Per-sequence tokens/s	Peak allocated GiB	Peak reserved GiB
32	4,128	129.0	18.94	19.49
64	6,685	104.5	20.16	21.80
128	9,049	70.7	26.74	27.92
256	11,231	43.9	39.90	41.77
320	11,074	34.6	46.49	50.69
384	12,099	31.5	53.06	57.27
512	12,497	24.4	61.26	66.97

Batch size	Decode tokens/s	Per-sequence tokens/s	Peak allocated GiB
1	230	229.6	14.27
2	358	178.8	14.27
4	692	172.9	14.27
8	1,322	165.2	14.40
16	2,442	152.6	15.21
32	4,053	126.7	16.84
64	6,698	104.7	20.10
128	9,066	70.8	26.60
256	11,100	43.4	39.62
320	11,062	34.6	46.14
384	12,086	31.5	52.64
512	12,481	24.4	60.71

History component	Role	Capacity
Recurrent state	Summarize the evolving context	Fixed-dimensional matrices
Local attention	Resolve recent token interactions	1,024 positions
RMA anchor bank	Retain individually retrievable distant records	1,024 records per KV group
Pending candidates	Collect records for the next admission update	256 positions

Adaptation stage	Query accuracy (%)	All four correct (%)
RMA adaptation (M0)	79.4	54.7
Joint adaptation (R)	96.1	87.9
Supervised fine-tuning (S)	93.8	82.4
Reinforcement learning (RLVR)	94.1	80.1

Final-model setting	Correct queries / total	Query accuracy (%)	Complete contexts / total	All four correct (%)
RMA enabled	964 / 1,024	94.1	205 / 256	80.1
RMA disabled	4 / 1,024	0.4	0 / 256	0.0

Task	Complete contexts / total	All four correct (%)
Exact retrieval	58 / 64	90.6
State-update interpretation	56 / 64	87.5
Cross-document combination	57 / 64	89.1
Variable binding	34 / 64	53.1
All tasks	205 / 256	80.1

Change to distant evidence	Correct pairs / total	Pair accuracy (%)
Answer should change	230 / 256	89.8
Answer should stay the same	247 / 256	96.5

Model	DROP F1	MuSR	LogiQA2
EdgeRock-4B	77.2	60.4	72.2
ARWKV-R1 1.5B	5.6	8.6	17.5
RWKV-7 G1i 1.5B	23.0	19.3	38.7
RWKV-7 G1j 1.5B	24.9	29.1	36.1
RWKV-7 G1i 2.9B	24.6	31.9	47.6
RWKV-7 G1j 2.9B	30.7	34.5	45.4
AHN-GDN Qwen2.5 3B	32.8	37.8	50.1
ARWKV-R1 7B	11.1	15.4	26.8
Falcon3-Mamba 7B Instruct	19.9	29.1	41.4
QRWKV7 7B Instruct	21.2	2.6	33.2
xLSTM 7B	29.0	19.3	20.0
RWKV-7 G1i 7.2B	33.3	49.6	53.1
RWKV-7 G1j 7.2B	34.2	48.6	56.4
RecurrentGemma 9B IT	37.1	35.2	40.2
RWKV-7 G1j 13.3B	51.5	51.4	66.6
AHN-GDN Qwen2.5 14B	66.5	56.4	72.3
Phi-3 Medium 4K Instruct	51.5	51.1	62.7

Model	HumanEval	MBPP	LiveCodeBench v6	IFEval	IFBench
EdgeRock-4B	96.7	87.8	61.6	88.0	65.3
Qwen3.5-4B	86.0	71.0	75.7	89.6	60.3
Gemma4-E2B-it	90.9	79.6	59.6	82.4	29.0
MiniCPM5-2B	97.6	94.8	83.3	86.3	64.0
LFM2.5-2.6B	92.9	73.6	59.7	93.3	56.2
Qwen3.5-2B	54.9	35.4	25.0	81.5	43.7
Granite4.2-3B	97.0	92.6	75.9	92.2	74.3
Phi-4-mini-reasoning	79.9	64.0	35.3	45.7	14.3
Gemma4-E4B-it	93.9	79.2	69.1	84.3	34.7
Nemotron-3-Nano-4B	90.9	72.2	66.4	89.5	57.7
Granite4.2-8B	96.3	95.4	85.6	94.6	77.0
MiniCPM4.1-8B	98.2	92.2	70.4	73.4	23.7

EdgeRock

Rock Matrix: toward Ultra-Fast Token

Quick start

Model directory

Original model and compact PyTorch

MLX

GGUF and ONNX

Mobile development previews

Reasoning and multimodal capability

Strong reasoning at constant memory

Comparison with compact models

Backbone capability retention

EdgeRock-4B vs Qwen3.5-4B: all 29 benchmarks

Visual reasoning

Inference efficiency

Constant history and GPU memory

Long-generation throughput

Best-batch output throughput

Single-request decoding

Concurrency scaling

Short-context decoding

How Rock Memory Anchors work

Distant-memory retrieval

Full benchmark results

Mathematics: bounded-state models

Knowledge and science: bounded-state models

Commonsense and reasoning: bounded-state models

Code and instruction following: bounded-state models

Reading and multi-step reasoning: bounded-state models

Mathematics: EdgeRock vs growing-history models

Knowledge and science: EdgeRock vs growing-history models

Commonsense and reasoning: EdgeRock vs growing-history models

Code and instruction following: EdgeRock vs growing-history models

Reading and multi-step reasoning: EdgeRock vs growing-history models

Documentation

What is EdgeRock-4B?

What stays constant in O(1) memory?

What do memory anchors add to sliding-window attention?

Does EdgeRock require quantization or speculative decoding?

Citation

License