MiniMax-Text-01
A Shanghai startup's January 2025 open-weight model with a 4 million-token context window — the longest publicly available at the time, twice Kimi's 2M and four times Gemini 1.5 Pro's 1M. Achieved this by replacing most layers' standard attention with "Lightning Attention" — a linear-time variant — and keeping standard attention only every 8th layer to preserve quality on short-context tasks.
How are Intelligence, Speed & Cost bucketed?
- Top 1%≤ 1%
- Top 5%≤ 5%
- Top 10%≤ 10%
- Good≤ 25%
- Medium≤ 50%
- Below avg> 50%
- Top 1%≥ 345 tok/s
- Top 5%≥ 237 tok/s
- Top 10%≥ 196 tok/s
- Good≥ 146 tok/s
- Medium≥ 90 tok/s
- Slow< 90 tok/s
- Freeopen weights · self-host
- Low< $1 / M out
- Moderate$1–5 / M out
- High≥ $5 / M out
Why it matters
MiniMax-Text-01 is the proof point that >>1M context isn't just a Google-DeepMind-scale-only feature — it's achievable with architectural changes that any frontier lab can adopt. The "context arms race" of 2024-25 has converged on hybrid linear / standard attention as the default approach, with MiniMax as one of the early demonstrations.
Core Capabilities
Context Window
Availability
Pricing Model
Capability / Performance
Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).
What it feels like
- Language model from MiniMax — see the linked sources below for benchmark and review coverage
- Tool-use and agent loops are the typical fit per the published model card
Best use cases
- Agent / tool-use workflows that match the model's published benchmarks
- See the model spec and sources block for benchmarked use cases
Tools to try
Not ideal for
- Tasks far outside the modalities listed in this model's spec
- Workflows where a more recent successor in the same family scores higher