g factor technologies Engineering Blog
g-ftech.com

Deep-dive systems research, post-training architectures, and hardware benchmarks from g factor technologies.

Multi-Reward RL, Part 2: Benchmarking GRPO, DAPO, and CISPO on Unseen Tasks

g factor technologies9 min read
Read on Scour →
Feeds

The Specialized Agent Shift: How Open Models Are Beating Frontier APIs in Production

g factor technologies10 min read
Read on Scour →
Feeds

The Verifier Design Playbook: How to Build RLVR Gyms That Models Can't Game

g factor technologies9 min read
Read on Scour →
Feeds

Long-Horizon Agents: Why Multi-Turn Reasoning Breaks and the Practical Training Tricks That Fix It

g factor technologies10 min read
Read on Scour →
Feeds

We Tried ISO-AdamW. AdamW Kept Its Job.

g factor technologies9 min read
Read on Scour →
Feeds

LoRA & DoRA: The Math, Memory, and Trade-offs

g factor technologies8 min read
Read on Scour →
Feeds

Physical AI: Why the Next Big Frontier Is Giving Software Agents Hands

g factor technologies9 min read
Read on Scour →
Feeds

Multi-Reward Reinforcement Learning for LLM Agents: Comparing PPO, GRPO, DAPO, and GDPO

g factor technologies7 min read
Read on Scour →
Feeds

From RLHF to RLVR: The Evolution of Reward Signals and the Battle Against Reward Hacking

g factor technologies8 min read
Read on Scour →
Feeds

Distributed Training & Inference: From CPUs and GPUs to a Cluster

g factor technologies10 min read
Read on Scour →
Feeds

AsyncGRPO: Eliminating GPU Idle Bubbles in Environment-Heavy RL Post-Training

g factor technologies8 min read
Read on Scour →
Feeds

Owning vs. Renting Intelligence: Why Enterprises Are Building Sovereign AI

g factor technologies7 min read
Read on Scour →
Feeds

Latent-GRPO: Reinforcement Learning in Continuous Thought Space

g factor technologies8 min read
Read on Scour →
Feeds

SFT vs. RL: What Changes Inside the Model?

g factor technologies6 min read
Read on Scour →
Feeds

The Limits of AI: Induction, Deduction, and Why Models Can't Jump

g factor technologies9 min read
Read on Scour →
Feeds

Benchmarking Qwen 3.8 27B Across Inference Providers: Together, Fireworks, Nebius, and g factor

g factor technologies10 min read
Read on Scour →
Feeds

vLLM Architecture, Memory and Benchmarks Deep Dive

g factor technologiesHN10 min read
Read on Scour →
Feeds

Keyboard shortcuts

Navigation

Next / previous post
j/k
Open post
oorEnter
Preview post
v

Post actions

Love post
a
Like post
l
Dislike
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Discover
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help