Llama 3.2 90B Instruct vs Mercury 3
Mercury 3 leads the LLM Stats Score 36.1 to 4.7. Mercury 3 is 2.2x cheaper per token.
Meta · Inception · Updated for 2026
Which is better?
Mercury 3 leads the overall LLM Stats Score 36.1 to 4.7, ranking #98 overall.
In the 1 individual benchmarks reported for both models, Mercury 3 wins 1; this is a narrower head-to-head signal than the composite indexes.
On price, Mercury 3 is roughly 2.2x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.
Mercury 3 also accepts a larger context window (1,000,000 input tokens), making it the stronger choice for long documents and large codebases.
Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.
Choose Llama 3.2 90B Instruct
- you need open weights you can self-host or fine-tune
Choose Mercury 3
- overall performance matters — it scores 36.1 and ranks #98 on LLM Stats
- your work emphasizes reasoning — it leads those capability indexes
- you value its reported benchmark strengths — it wins 1 of 1 exact shared results
- cost matters — it's about 2.2x cheaper per token
- you process long inputs — it offers a 1,000,000 token context window
- you want the most recent training data — it shipped Oct 2026
At a glance
The differences that matter most.
Individual benchmarks
13 reported for Llama 3.2 90B Instruct · 12 for Mercury 3
Llama 3.2 90B Instruct outperforms in 0 benchmarks, while Mercury 3 is better at 1 benchmark (GPQA).
Mercury 3 significantly outperforms across most benchmarks.
Human preference
Blind head-to-head votes and playground preference scores
Pricing Analysis
Price comparison per million tokens
For input processing, Llama 3.2 90B Instruct ($0.35/1M tokens) is 3.5x more expensive than Mercury 3 ($0.10/1M tokens).
For output processing, Llama 3.2 90B Instruct ($0.40/1M tokens) is 1.1x more expensive than Mercury 3 ($0.35/1M tokens).
In conclusion, Llama 3.2 90B Instruct is more expensive than Mercury 3.*
* Using a 3:1 ratio of input to output tokens
Context Window
Maximum input and output token capacity
Mercury 3 accepts 1,000,000 input tokens compared to Llama 3.2 90B Instruct's 128,000 tokens. Mercury 3 can generate longer responses up to 131,072 tokens, while Llama 3.2 90B Instruct is limited to 128,000 tokens.
Input capabilities
Documented input modalities across available providers
Llama 3.2 90B Instruct supports multimodal inputs, whereas Mercury 3 does not.
Llama 3.2 90B Instruct can handle both text and other forms of data like images, making it suitable for multimodal applications.
Llama 3.2 90B Instruct
Mercury 3
License
Usage and distribution terms
Llama 3.2 90B Instruct is licensed under Llama 3.2, while Mercury 3 uses a proprietary license.
License differences may affect how you can use these models in commercial or open-source projects.
Llama 3.2
Open weights
Proprietary
Closed source
Release Timeline
When each model was launched
Llama 3.2 90B Instruct was released on 2024-09-25, while Mercury 3 was released on 2026-10-19.
Mercury 3 is 25 months newer than Llama 3.2 90B Instruct.
Sep 25, 2024
2.1 years ago
Oct 19, 2026
0 days ago
2.1yr newerKnowledge Cutoff
When training data ends
Neither model specifies a knowledge cutoff date.
Unable to compare the recency of their training data.
Provider Availability
Llama 3.2 90B Instruct is available from DeepInfra, Bedrock, Fireworks, Together, Hyperbolic. Mercury 3 is available from Inception.
Llama 3.2 90B Instruct
Mercury 3
Outputs Comparison
Judge for yourself.
Run your own prompts against Llama 3.2 90B Instruct and Mercury 3 side-by-side, then vote on the output you prefer.
FAQ
Common questions about Llama 3.2 90B Instruct vs Mercury 3.