Muhammadreza

OpenMythos-9B-1M-heretic

Deploy Dedicated

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Model Details

Model Provider

Muhammadreza

Model Tree

Base

Qwen/Qwen3.5-9B

Fine-tuned

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

First Benchmark

Table with columns: Tasks, Version, Filter, n-shot, Metric, Value, Stderr
Tasks	Version	Filter	n-shot	Metric		Value		Stderr
arc_challenge	1.0	none	0	acc	↑	0.5360	±	0.0223
		none	0	acc_norm	↑	0.5580	±	0.0222
gpqa_diamond_zeroshot	2.2	none	0	acc	↑	0.4495	±	0.0354
		none	0	acc_norm	↑	0.4495	±	0.0354
gsm8k	3.0	flexible-extract	5	exact_match	↑	0.8300	±	0.0168
		strict-match	5	exact_match	↑	0.8340	±	0.0167
gsm8k_cot	3.0	flexible-extract	8	exact_match	↑	0.8440	±	0.0162
		strict-match	8	exact_match	↑	0.7560	±	0.0192

Detailed Benchmark

Table with columns: Tasks, Version, Filter, n-shot, Metric, Value, Stderr
Tasks	Version	Filter	n-shot	Metric		Value		Stderr
stem	2.0	none	0	acc	↑	0.7707	±	0.0072
- abstract_algebra	1.0	none