The AI arena is free today

Open Superagent

Granite 3.3 8B Instruct vs Kimi K2.8 Preview

Comparing Granite 3.3 8B Instruct and Kimi K2.8 Preview across benchmarks, pricing, and capabilities.

IBM · Moonshot AI · Updated for 2026

Which is better?

Granite 3.3 8B Instruct and Kimi K2.8 Preview trade strengths across price, capabilities, and technical limits. The better choice depends on the workload.

Kimi K2.8 Preview also accepts a larger context window (1,048,576 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 Granite 3.3 8B Instruct

  • you need open weights you can self-host or fine-tune

Choose Kimi K2.8 Preview

  • you process long inputs — it offers a 1,048,576 token context window
  • you want the most recent training data — it shipped Sep 2026

At a glance

The differences that matter most.

Granite 3.3 8B Instruct
IBM
Kimi K2.8 Preview
Moonshot AI
Benchmark wins
—
—
Input price
$0.50 / M
— / M
Output price
$0.50 / M
— / M
Context window
128,000
1,048,576

Individual benchmarks

14 reported for Granite 3.3 8B Instruct · 0 for Kimi K2.8 Preview

No common benchmarks found

Granite 3.3 8B Instruct and Kimi K2.8 Previewdon't have any common benchmark datasets to compare. They may have been evaluated on different testing suites.

Human preference

Blind head-to-head votes and playground preference scores

Context Window

Maximum input and output token capacity

Kimi K2.8 Preview accepts 1,048,576 input tokens compared to Granite 3.3 8B Instruct's 128,000 tokens. Only Granite 3.3 8B Instruct specifies output context (8,192 tokens).

IBM
Granite 3.3 8B Instruct
Input128,000 tokens
Output8,192 tokens
Moonshot AI
Kimi K2.8 Preview
Input1,048,576 tokens
Output- tokens
Sat Sep 19 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Both Granite 3.3 8B Instruct and Kimi K2.8 Preview support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

Granite 3.3 8B Instruct

Text
Images
Audio
Video

Kimi K2.8 Preview

Text
Images
Audio
Video

License

Usage and distribution terms

Granite 3.3 8B Instruct is licensed under Apache 2.0, while Kimi K2.8 Preview uses a proprietary license.

License differences may affect how you can use these models in commercial or open-source projects.

Granite 3.3 8B Instruct

Apache 2.0

Open weights

Kimi K2.8 Preview

Proprietary

Closed source

Release Timeline

When each model was launched

Granite 3.3 8B Instruct was released on 2025-04-16, while Kimi K2.8 Preview was released on 2026-09-11.

Kimi K2.8 Preview is 17 months newer than Granite 3.3 8B Instruct.

Granite 3.3 8B Instruct

Apr 16, 2025

1.4 years ago

Kimi K2.8 Preview

Sep 11, 2026

1 weeks ago

1.4yr newer

Knowledge Cutoff

When training data ends

Granite 3.3 8B Instruct has a documented knowledge cutoff of 2024-04-01, while Kimi K2.8 Preview's cutoff date is not specified.

We can confirm Granite 3.3 8B Instruct's training data extends to 2024-04-01, but cannot make a direct comparison without Kimi K2.8 Preview's cutoff date.

Granite 3.3 8B Instruct

Apr 2024

Kimi K2.8 Preview

—

Provider Availability

Granite 3.3 8B Instruct is available from Replicate. Kimi K2.8 Preview is available from Moonshot AI.

Granite 3.3 8B Instruct

replicate logo
Replicate
Input Price:Input: $0.50/1MOutput Price:Output: $0.50/1M

Kimi K2.8 Preview

moonshot logo
Unknown Organization
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Judge for yourself.

Run your own prompts against Granite 3.3 8B Instruct and Kimi K2.8 Preview side-by-side, then vote on the output you prefer.

Granite 3.3 8B Instruct
✓ Preferred
Kimi K2.8 Preview
Open in Playground

FAQ

Common questions about Granite 3.3 8B Instruct vs Kimi K2.8 Preview.

Which is better, Granite 3.3 8B Instruct or Kimi K2.8 Preview?

Granite 3.3 8B Instruct (IBM) and Kimi K2.8 Preview (Moonshot AI) each have strengths in different areas. Compare their benchmark scores, pricing, context windows, and capabilities above to determine which fits your needs.

How does Granite 3.3 8B Instruct compare to Kimi K2.8 Preview in benchmarks?

Granite 3.3 8B Instruct scores HumanEval: 89.7%, AttaQ: 88.5%, HumanEval+: 86.1%, AIME 2024: 81.2%, GSM8k: 80.9%.

What are the context window sizes for Granite 3.3 8B Instruct and Kimi K2.8 Preview?

Granite 3.3 8B Instruct supports 128K tokens and Kimi K2.8 Preview supports 1.0M tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Granite 3.3 8B Instruct and Kimi K2.8 Preview?

Key differences include context window (128K vs 1.0M), licensing (Apache 2.0 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes Granite 3.3 8B Instruct and Kimi K2.8 Preview?

Granite 3.3 8B Instruct is developed by IBM and Kimi K2.8 Preview is developed by Moonshot AI.

Related comparisons