1stTop pick
Researched 36d agoLlama 4 Scout 17B-16E Instruct
- Context
- 10m
- Output (from)
- $0.220 / 1M
Best Long Context LLMs (2026)
Last refreshed 2026-07-11. Next refresh: weekly.
LLMs with the largest context windows in 2026 — from 128K to 2M tokens. Ranked by window size with pricing and retrieval accuracy.
New release-watch entry: Gemini 3.5 Pro is tracked with a 2M-token context window while it remains in preview.
Verdict
Use Llama 4 Scout 17B-16E Instruct for long-context work today.
MiniMax-01 is the runner-up: 10m vs 4m on Context.
2ndShortlist
Researched 21d agoMiniMax-01
- Context
- 4m
- Output (from)
- $1.10 / 1M
3rdShortlist
Researched 4d agoGPT-5.6 Sol
- Context
- 1.05m
- Output (from)
- $30.00 / 1M
How we rank
Long-context leaders sort by the widest shipped context window first, then recency — benchmarks matter, but literal window size is the gate for this category.
- Eligibility — Models with ≥128k tokens of declared context and not deprecated.
- Primary ranking — Larger parsed context wins; newer release on ties.
- Variant collapse — We keep one row per model family (`familySlug` + parameter tier). When headline scores tie within ±0.5 pt (±10 Elo on Chatbot Arena), we pick the canonical SKU by lowest tracked input price, then GA over preview or limited access, then newest `release`. A folded sibling within the benchmark noise band can show a "Tied within margin" chip on that score cell.
- Pricing — Lowest tracked provider rates where available.
| # | Model | Input $/1M | Output $/1M | |
|---|---|---|---|---|
| 1 | RWKV-7 Goose 2.9B | — | — | |
| 2 | RWKV-7 Goose 1.5B | — | — | |
| 3 | RWKV-7 Goose 0.4B | — | — | |
| 4 | RWKV-7 Goose 0.1B | — | — | |
| 5 | RWKV-6 Finch 14B | — | — | |
| 6 | RWKV-6 Finch 7B | — | — | |
| 7 | RWKV-6 Finch 3B | — | — | |
| 8 | RWKV-6 Finch 1.6B | — | — | |
| 9 | LTM-2-mini | — | — | |
| 10 | Llama 4 Scout 17B-16E Instruct Vision | $0.08 | $0.22 | |
| 11 | LTM-1 | — | — | |
| 12 | MiniMax-01 Vision | $0.20 | $1.10 | |
| 13 | Gemini 3.5 Pro PreviewReasoningVision | — | — | |
| 14 | Gemini 1.5 Pro 002 | — | — | |
| 15 | Gemini 1.5 Pro Experimental 0827 | — | — | |
| 16 | Gemini 1.5 Pro Experimental 0801 | — | — | |
| 17 | Gemini 1.5 Pro | $1.25 | $5.00 | |
| 18 | GPT-5.6 Sol ReasoningVisionTools | $5.00 | $30.00 | |
| 19 | GPT-5.6 Terra ReasoningVisionTools | $2.50 | $15.00 | |
| 20 | GPT-5.6 Luna ReasoningVisionTools | $1.00 | $6.00 | |
| 21 | GPT-5.5 ReasoningVisionTools | $5.00 | $30.00 | |
| 22 | GPT-5.5 Pro ReasoningVisionTools | $30.00 | $180.00 | |
| 23 | GPT-5.4 ReasoningVisionTools | $2.50 | $15.00 | |
| 24 | GPT-5.4 Pro ReasoningVisionTools | $30.00 | $180.00 | |
| 25 | Gemini 3.5 Flash ReasoningVisionTools | $1.50 | $9.00 | |
| 26 | Antigravity Agent PreviewReasoningVision | — | — | |
| 27 | Gemini 3.1 Flash-Lite VisionTools | $0.25 | $1.50 | |
| 28 | Xiaomi MiMo-V2.5-Pro Tools | $0.43 | $0.87 | |
| 29 | Xiaomi MiMo-V2.5 ReasoningVisionTools | $0.14 | $0.28 | |
| 30 | Nemotron-Cascade-2-30B-A3B | — | — | |
| 31 | MiMo-V2-Pro | $1.00 | $3.00 | |
| 32 | Nemotron 3 Super-120B-A12B | $0.09 | $0.45 | |
| 33 | Gemini 2.5 Pro Computer Use Preview PreviewVisionTools | $1.25 | $10.00 | |
| 34 | Llama 3 70B Gradient 1048K | — | — | |
| 35 | GPT-4.1 VisionTools | $2.00 | $8.00 | |
| 36 | GPT-4.1 Mini VisionTools | $0.40 | $1.60 | |
| 37 | Muse Spark 1.1 ReasoningVisionTools | $1.25 | $4.25 | |
| 38 | Claude Sonnet 5 ReasoningVisionTools | $2.00 | $10.00 | |
| 39 | LongCat-2.0 ReasoningTools | $0.75 | $2.95 | |
| 40 | GLM-5.2 ReasoningTools | $1.40 | $4.40 |
Honorable mentions
Next seats in this ranking. Lines below are from each model's stored description in LLMReference seed data—spot-check the model page before relying on a capability claim.
OpenAI's balanced mid-tier GPT-5.6 model. Terra delivers performance competitive with GPT-5.5 at about 2x lower cost, suited for high-volume production workloads such as customer support, document analysis, and internal tooling. Generally available July 9, 2026 across ChatGPT, Codex, and the OpenAI API (model ID gpt-5.6-terra). Supports text and image inputs, reasoning, tool use, prompt caching, and Batch API with a 1,050,000-token context window and 128K max output.
1.05m
Context
OpenAI's fast, low-cost GPT-5.6 model. Luna is the affordable tier in the Sol, Terra, and Luna lineup, optimized for latency-sensitive applications such as summarization, drafting, autocomplete, and routine automation. Generally available July 9, 2026 across ChatGPT, Codex, and the OpenAI API (model ID gpt-5.6-luna). Supports text and image inputs, reasoning, tool use, prompt caching, and Batch API with a 1,050,000-token context window and 128K max output.
1.05m
Context
- #6GPT-5.5
GPT-5.5 is OpenAI's fully retrained agentic model, released April 23, 2026. Optimised for agentic coding, computer use, knowledge work, and early scientific research. Achieves 82.7% on Terminal-Bench 2.0 (Codex CLI scaffold), 84.9% on GDPval, 58.6% on SWE-Bench Pro, 93.6% on GPQA Diamond, and 82.6% on SWE-Bench Verified (Vals.ai independent harness). Knowledge cutoff December 2025. Supports reasoning effort levels (none/low/medium/high/xhigh). Context window 1,050,000 tokens with a long-context surcharge above 272K tokens. Model ID: gpt-5.5.
1.05m
Context
Compare Top Picks
Side-by-side comparison of the top picks by price, benchmark, and API access.
Browse Other Categories
Best LLMs for Code GenerationBest LLMs for RAGBest AI Agent Models 2026: SWE-bench RankedBest LLMs for ClassificationBest Open Source LLMsBest Multimodal / Vision LLMsBest LLM for Translation in 2026Best AI Image Models in 2026Best AI Video Models in 2026Best LLMs for Reasoning & MathBest Small Language Models (SLMs)Best LLMs for Function Calling & Tool UseCheapest LLM APIs You Can Call Right NowBest Mainstream LLM APIs, RankedBest LLMs for EnterpriseBest Free LLMs You Can Use Right NowBest LLMs for WritingBest LLMs for MarketingBest LLMs for Customer Support
Frequently asked questions
Which LLM is best for long-context work?
Llama 4 Scout 17B-16E Instruct is the current LLMReference top pick for long-context work. The verdict uses the stored category signal Context: 10m. Output pricing starts at $0.22 per 1M tokens. Review the linked model and provider pages before production use because availability and pricing can change.
How does Llama 4 Scout 17B-16E Instruct compare to MiniMax-01 for long-context work?
Llama 4 Scout 17B-16E Instruct leads MiniMax-01 in the visible shortlist on Context: 10m versus 4m. The pricing cards show Llama 4 Scout 17B-16E Instruct: output pricing starts at $0.22 per 1m tokens and MiniMax-01: output pricing starts at $1.10 per 1m tokens.
How does LLMReference rank LLMs for long-context work?
LLMReference ranks LLMs for long-context work from stored model, benchmark, freshness, and pricing data. The current methodology summary is: Long-context leaders sort by the widest shipped context window first, then recency — benchmarks matter, but literal window size is the gate for this category.
How often is this list updated?
The LLM rankings on this page are updated daily as new benchmark scores, provider availability, and pricing data are tracked. The "as of" date at the top of the page shows the most recent refresh.
How do you decide which models appear in the top 3?
The podium picks are driven by the primary benchmark signal for this category (shown in the Methodology section), filtered to non-deprecated models with confirmed API availability. In ties, we prefer the more recently released model.
Are preview or beta models included?
Preview models appear in the "Watch list" section but are not in the main ranked podium unless the category explicitly allows it (e.g., /best/coding and /best/agents, where preview models often lead benchmarks).
Can I compare two specific models head-to-head?
Yes — use the Compare tool at llmreference.com/compare for a side-by-side breakdown of context window, pricing, benchmarks, and provider availability.
Is the pricing data real-time?
Pricing is tracked from provider documentation and updated regularly. It reflects the best available public data, not live API quotes — always verify before billing.