InternLM 2.5 20B vs Cohere Rerank 3.5

Compare InternLM 2.5 20B and Cohere Rerank 3.5: pricing, performance, context window, latency, and best use cases. Side-by-side comparison on XALEN.

Updated 2026-05-21 · By Abhishek Raj · Our methodology

All Shanghai AI Lab models All Cohere models What is an LLM API? Python Quickstart What is inference?
Feature InternLM 2.5 20B Cohere Rerank 3.5
CategoryOpen SourceReranking
Parameters20B~600M
Context Window256K4K
Input Price$0.02/1M tokens$0.002/search/1M tokens
Output Price$0.04/1M tokensN/A/1M tokens
Latency~180ms~25ms

Choose InternLM 2.5 20B when:

  • ✓ Long context tasks
  • ✓ Research
  • ✓ Multilingual
Key Strengths:

256K context, Strong reasoning, Good multilingual

Choose Cohere Rerank 3.5 when:

  • ✓ Search reranking
  • ✓ RAG improvement
  • ✓ Result quality
Key Strengths:

Higher quality, Multilingual, Fast

Verdict: InternLM 2.5 20B vs Cohere Rerank 3.5

For cost efficiency, Cohere Rerank 3.5 wins at $0.002/search/1M input tokens. For speed, InternLM 2.5 20B is faster at ~180ms. InternLM 2.5 20B excels at Long context tasks while Cohere Rerank 3.5 is better for Search reranking. Both are available on XALEN through a single API — try them in the Playground to see which fits your workload.

Detailed Analysis

Pricing Comparison

InternLM 2.5 20B costs $0.02/1M input tokens and $0.04/1M output tokens. Cohere Rerank 3.5 costs $0.002/search input and N/A output. Cohere Rerank 3.5 is 10.0x cheaper on input tokens. XALEN offers batch processing at 50% discount on both models.

Performance & Context

InternLM 2.5 20B has a 256K context window with ~180ms latency. Cohere Rerank 3.5 offers 4K context at ~25ms. InternLM 2.5 20B has the larger context window.

Best For

InternLM 2.5 20B (Open Source) is optimized for: Long context tasks, Research, Multilingual. Cohere Rerank 3.5 (Reranking) works best for: Search reranking, RAG improvement, Result quality.

Try Both on XALEN

Both models are available through XALEN's OpenAI-compatible API. Switch between them by changing the model parameter:

from xalen import XALEN

client = XALEN(api_key="xln_test_YOUR_KEY")

# Use InternLM 2.5 20B
response_a = client.chat.completions.create(
    model="internlm-2-5-20b",
    messages=[{"role": "user", "content": "Your question here"}]
)

# Use Cohere Rerank 3.5
response_b = client.chat.completions.create(
    model="rerank-v3-5",
    messages=[{"role": "user", "content": "Your question here"}]
)

Start Building with XALEN

200+ AI models. One API. Pay-as-you-go.

Get API Key Try in Playground

Frequently Asked Questions

Which is better, InternLM 2.5 20B or Cohere Rerank 3.5?

InternLM 2.5 20B (Open Source, 20B) offers 256K context. Cohere Rerank 3.5 (Reranking, ~600M) offers Higher quality. Choose InternLM 2.5 20B for Long context tasks or Cohere Rerank 3.5 for Search reranking.

How much does InternLM 2.5 20B cost vs Cohere Rerank 3.5?

InternLM 2.5 20B: $0.02/1M input, $0.04/1M output. Cohere Rerank 3.5: $0.002/search/1M input, N/A/1M output. Both available on XALEN with batch processing at 50% discount.

Can I use both models on XALEN?

Yes. XALEN provides 200+ models through a single OpenAI-compatible API. Switch between InternLM 2.5 20B and Cohere Rerank 3.5 by changing the model parameter. No code changes needed.

Related Comparisons

InternLM 2.5 20B vs Gemma 3 27B InternLM 2.5 20B vs Llama 4 Scout InternLM 2.5 20B vs Llama 4 Maverick InternLM 2.5 20B vs Llama 3.3 70B InternLM 2.5 20B vs Llama 3.1 405B InternLM 2.5 20B vs Llama 3.1 70B Turbo

Last updated: 2026-05-21. Pricing and specifications may change. Check pricing page for latest rates.