Catalog data · Updated August 12, 2026

Largest Context Window LLMs

Compare active text-generation model identities with recorded context windows of at least 128,000 tokens. Large outlier values are excluded from the ranking.

Objective catalog fields No overall quality score
546
Long-context identities
25
Providers shown
1
Data sections
6,429
Catalog records

Largest context windows

Active text-generation identities from 128,000 through 10 million tokens.

Showing 50 of 546

Model Provider offers Why listed Context Input / output price Updated
DeepSeek-R1
DeepSeek-R1
Togetherai
1 offer
163839 token context 163,839
$3.00 / $7.00
per 1M tokens, when known
March 24, 2025
AionLabs: Aion-2.0
aion-2.0
Openrouter
1 offer
131072 token context 131,072
$0.80 / $1.60
per 1M tokens, when known
February 23, 2026
AionLabs: Aion-3.0
aion-3.0
Openrouter
1 offer
131072 token context 131,072
$3.00 / $6.00
per 1M tokens, when known
July 7, 2026
AionLabs: Aion-3.0-Mini
aion-3.0-mini
Openrouter
1 offer
131072 token context 131,072
$0.70 / $1.40
per 1M tokens, when known
July 7, 2026
Arcee AI: Virtuoso Large
virtuoso-large
Openrouter
1 offer
131072 token context 131,072
$0.75 / $1.20
per 1M tokens, when known
May 5, 2025
Compound
groq/compound
Groq
1 offer
131072 token context 131,072
N/A / N/A
per 1M tokens, when known
September 4, 2025
Compound Mini
compound-mini
Groq
1 offer
131072 token context 131,072
N/A / N/A
per 1M tokens, when known
September 4, 2025
Deep Research Max Preview (Apr-21-2026)
deep-research-max-preview-04-2026
Google
1 offer
131072 token context 131,072
$2.00 / $12.00
per 1M tokens, when known
April 21, 2026
Deep Research Preview (Apr-21-2026)
deep-research-preview-04-2026
Google
1 offer
131072 token context 131,072
$2.00 / $12.00
per 1M tokens, when known
April 21, 2026
DeepSeek V3.1
DeepSeek-V3-1
Togetherai
1 offer
131072 token context 131,072
$0.60 / $1.70
per 1M tokens, when known
August 21, 2025
DeepSeek-V3
DeepSeek-V3
Togetherai
1 offer
131072 token context 131,072
$1.25 / $1.25
per 1M tokens, when known
May 29, 2025
Gemini 2.5 Computer Use Preview 10-2025
gemini-2.5-computer-use-preview-10-2025
Google
1 offer
131072 token context 131,072
$1.25 / $10.00
per 1M tokens, when known
October 7, 2025
Gemini Robotics-ER 1.6 Preview
gemini-robotics-er-1.6-preview
Google
1 offer
131072 token context 131,072
$1.00 / $5.00
per 1M tokens, when known
April 14, 2026
Gemma 4 31B IT
gemma-4-31b
Cerebras
1 offer
131072 token context 131,072
$0.99 / $1.49
per 1M tokens, when known
July 1, 2026
GLM-4.5-Flash
glm-4.5-flash
Zai
1 offer
131072 token context 131,072
$0.00 / $0.00
per 1M tokens, when known
July 28, 2025
Google: Gemma 3 12B
google/gemma-3-12b-it
Openrouter
1 offer
131072 token context 131,072
$0.05 / $0.15
per 1M tokens, when known
March 13, 2025
Google: Gemma 3 4B
google/gemma-3-4b-it
Openrouter
1 offer
131072 token context 131,072
$0.05 / $0.10
per 1M tokens, when known
March 13, 2025
GPT OSS 120B
gpt-oss-120b
Cerebras, Cloudflare workers ai, Fireworks ai +4
7 offers
131072 token context 131,072
$0.04 / $0.17
per 1M tokens, when known
June 16, 2026
GPT OSS 20B
gpt-oss-20b
Cloudflare workers ai, Fireworks ai, Groq +2
5 offers
131072 token context 131,072
$0.03 / $0.13
per 1M tokens, when known
September 25, 2025
gpt-oss:120b
gpt-oss:120b
Ollama cloud
1 offer
131072 token context 131,072
N/A / N/A
per 1M tokens, when known
January 19, 2026
gpt-oss:20b
gpt-oss:20b
Ollama cloud
1 offer
131072 token context 131,072
N/A / N/A
per 1M tokens, when known
January 19, 2026
IBM: Granite 4.1 8B
granite-4.1-8b
Openrouter
1 offer
131072 token context 131,072
$0.05 / $0.10
per 1M tokens, when known
April 30, 2026
Llama 3.1 8B
llama-3.1-8b-instant
Groq
1 offer
131072 token context 131,072
$0.05 / $0.08
per 1M tokens, when known
July 23, 2024
Llama 3.3 70B
llama-3.3-70b-versatile
Groq
1 offer
131072 token context 131,072
$0.59 / $0.79
per 1M tokens, when known
December 6, 2024
Llama 3.3 70B
meta-llama/Llama-3.3-70B-Instruct-Turbo
Togetherai
1 offer
131072 token context 131,072
$1.04 / $1.04
per 1M tokens, when known
July 2, 2026
Llama Guard 3 8B
llama-guard-3-8b
Cloudflare workers ai
1 offer
131072 token context 131,072
$0.48 / $0.03
per 1M tokens, when known
January 22, 2025
Meta: Llama 3.1 70B Instruct
llama-3.1-70b-instruct
Openrouter
1 offer
131072 token context 131,072
$0.40 / $0.40
per 1M tokens, when known
July 23, 2024
Meta: Llama 3.1 8B Instruct
llama-3.1-8b-instruct
Openrouter
1 offer
131072 token context 131,072
$0.05 / $0.08
per 1M tokens, when known
July 23, 2024
Meta: Llama 3.2 3B Instruct
llama-3.2-3b-instruct
Openrouter
1 offer
131072 token context 131,072
$0.05 / $0.33
per 1M tokens, when known
September 25, 2024
Meta: Llama 3.3 70B Instruct
llama-3.3-70b-instruct
Openrouter
1 offer
131072 token context 131,072
$0.10 / $0.32
per 1M tokens, when known
December 6, 2024
Mistral Large 2.1
mistral-large-2411
Mistral
1 offer
131072 token context 131,072
$2.00 / $6.00
per 1M tokens, when known
November 18, 2024
Mistral Large 2407
mistral-large-2407
Openrouter
1 offer
131072 token context 131,072
$2.00 / $6.00
per 1M tokens, when known
November 19, 2024
Mistral Medium 3
mistral-medium-2505
Mistral
1 offer
131072 token context 131,072
$0.40 / $2.00
per 1M tokens, when known
May 7, 2025
Mistral: Ministral 3 3B 2512
mistralai/ministral-3b-2512
Openrouter
1 offer
131072 token context 131,072
$0.10 / $0.10
per 1M tokens, when known
December 2, 2025
Mistral: Mistral Medium 3
mistral-medium-3
Openrouter
1 offer
131072 token context 131,072
$0.40 / $2.00
per 1M tokens, when known
May 7, 2025
Mistral: Mistral Medium 3.1
mistral-medium-3.1
Openrouter
1 offer
131072 token context 131,072
$0.40 / $2.00
per 1M tokens, when known
August 13, 2025
Mistral: Mistral Nemo
mistral-nemo
Mistral, Openrouter
2 offers
131072 token context 131,072
$0.02 / $0.03
per 1M tokens, when known
July 1, 2024
MoonshotAI: Kimi K2 0711
moonshotai/kimi-k2
Openrouter
1 offer
131072 token context 131,072
$0.57 / $2.30
per 1M tokens, when known
July 11, 2025
Nano Banana 2
gemini-3.1-flash-image
Google
1 offer
131072 token context 131,072
$0.50 / $3.00
per 1M tokens, when known
unknown
Nous: Hermes 3 405B Instruct
nousresearch/hermes-3-llama-3.1-405b
Openrouter
1 offer
131072 token context 131,072
$1.00 / $1.00
per 1M tokens, when known
August 16, 2024
Nous: Hermes 3 70B Instruct
nousresearch/hermes-3-llama-3.1-70b
Openrouter
1 offer
131072 token context 131,072
$0.70 / $0.70
per 1M tokens, when known
August 18, 2024
Nous: Hermes 4 405B
hermes-4-405b
Openrouter
1 offer
131072 token context 131,072
$1.00 / $3.00
per 1M tokens, when known
August 26, 2025
Nous: Hermes 4 70B
hermes-4-70b
Openrouter
1 offer
131072 token context 131,072
$0.13 / $0.40
per 1M tokens, when known
August 26, 2025
OpenAI: gpt-oss-20b (free)
gpt-oss-20b:free
Openrouter
1 offer
131072 token context 131,072
$0.00 / $0.00
per 1M tokens, when known
August 5, 2025
OpenAI: gpt-oss-safeguard-20b
gpt-oss-safeguard-20b
Openrouter
1 offer
131072 token context 131,072
$0.08 / $0.30
per 1M tokens, when known
October 29, 2025
QVQ Max
qvq-max
Alibaba
1 offer
131072 token context 131,072
$1.20 / $4.80
per 1M tokens, when known
March 25, 2025
Qwen-VL Max
qwen-vl-max
Alibaba
1 offer
131072 token context 131,072
$0.80 / $3.20
per 1M tokens, when known
August 13, 2025
Qwen-VL Plus
qwen-vl-plus
Alibaba
1 offer
131072 token context 131,072
$0.21 / $0.63
per 1M tokens, when known
August 15, 2025
Qwen2.5 14B Instruct
qwen2-5-14b-instruct
Alibaba
1 offer
131072 token context 131,072
$0.35 / $1.40
per 1M tokens, when known
unknown
Qwen2.5 32B Instruct
qwen2-5-32b-instruct
Alibaba
1 offer
131072 token context 131,072
$0.70 / $2.80
per 1M tokens, when known
unknown

How the context ranking works

The page orders active text-generation identities by their largest recorded provider context limit.

Inclusion criteria

  • The model identity has at least one offer that passes the active text-generation rules.
  • The largest recorded context window is from 128,000 through 10 million tokens.

Exclusions

  • The list excludes missing and nonnumeric context values.
  • The list excludes values above 10 million tokens as unverified outliers.

Data rules

  • Context value: A grouped identity uses the largest recorded context limit among its eligible provider offers.
  • Ordering: Model identities are ordered by context size, then name and model ID for stable output.

Limits of the data

  • A large context window does not prove strong recall, reasoning, speed, or low cost across the full window.
  • Providers can apply different context limits to the same model identity.

Read context limits with care

The advertised limit is only one part of long-context performance. Test retrieval quality, latency, and total prompt cost with data that matches your application.

Sources

  • llm_db — Source for active offers and recorded model context limits. Checked 2026-07-30.