How much does Magnum v4 72B cost?

Magnum v4 72B costs $3.00 per million input tokens and $5.00 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.011 per message.

What benchmarks has Magnum v4 72B been tested on?

Magnum v4 72B has been evaluated on 6 benchmarks. Top scores: IFEval: 56.3, BBH (HuggingFace): 35.5, MMLU-PRO: 31.4.

Is Magnum v4 72B open source?

Yes, Magnum v4 72B is open source.

How does Magnum v4 72B compare to Phi 3 Mini 4k Instruct?

Magnum v4 72B has an average score of 51.2 while Phi 3 Mini 4k Instruct scores 51.1. Magnum v4 72B outperforms Phi 3 Mini 4k Instruct overall. See full comparison →

Home/Models/Magnum v4 72B

Magnum v4 72B

Name: Magnum v4 72B
Price: 3 USD
Author: anthracite-org

by anthracite-org · Released Oct 2024

Open Source

51.2

avg score

Rank #103

Compare

Better than 56% of all models

Context

16K tokens (~8 books)

Input $/1M

$3.00

Output $/1M

$5.00

Type

text

License

Open Source

Benchmarks

6 tested

Data updated today

About

This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus). The model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-

Tested on 6 benchmarks with 27.9% average. Top scores: IFEval (56.3%), BBH (HuggingFace) (35.5%), MMLU-PRO (31.4%).

Looking for similar performance at lower cost?
Qwen3 32B scores 51.7 (101% as good) at $0.08/1M input · 97% cheaper

Capabilities

reasoning

13.4

#118 globally

math

20.0

#160 globally

knowledge

20.9

#197 globally

language

56.3

#96 globally

general

35.5

#26 globally

Benchmark Scores

Compare All

Tested on 6 benchmarks · Ranked across 5 categories

Score Distribution (all 233 models)

0255075100

▲ You are here

reasoningCompare reasoning →

MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

13.4—

mathCompare math →

MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

20.0—

knowledgeCompare knowledge →

MMLU-PRO

HuggingFace MMLU-Pro. Harder version of MMLU with 10 answer choices instead of 4 and more challenging questions.

31.4—

GPQA

HuggingFace evaluation of GPQA (Graduate-Level Google-Proof Q&A). PhD-level science questions that cannot be easily searched.

10.4—

Quick compare:

vs Phi 3 Mini 4k Instruct

vs Phi 3.5 Mini Instruct

vs Grok 4 Fast

Excellent (85+) Good (70-85) Average (50-70) Below (<50)

Similar Models

Phi 3 Mini 4k Instruct

Microsoft

51.1TBD

Phi 3.5 Mini Instruct

Links

Info

anthracite-org Pricing explorer Developers · API

Research

Documentation

Community

Source Code

BenchGecko API

magnum-v4-72b

Specifications

Typetext
Context16K tokens (~8 books)
ReleasedOct 2024
LicenseOpen Source
StatusActive
Cost / Message~$0.011

Available On

anthracite-org$3.00

Frequently Asked Questions

Magnum v4 72B is an open-source text AI model by anthracite-org, released in October 2024. It has an average benchmark score of 51.2. Context window: 16K tokens.

Benchmarks

IFEval BBH (HuggingFace)MMLU-PRO MATH Level 5 MUSR

anthracite-org · Provider anthracite-org · Economy All Models Compare Models Pricing Developers · API

Magnum v4 72B

Frequently Asked Questions

Related Models

Benchmarks

Related Pages