Trinity Large Thinking — Hardware Requirements & GPU Compatibility
ChatReasoning
Trinity Large Thinking is a 398.6B-parameter open language model from Arcee AI. It supports a context window of up to 262,144 tokens. At BF16 it needs about 797.82 GB of VRAM — see which GPUs and Macs can run it below.
Trinity Large Thinking runs on a multi-GPU server
BF16 needs ~799.6 GB · about 7× H200 SXM 141 GB
Other
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Trinity Large Thinking
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve arcee-ai/Trinity-Large-Thinking --max-model-len 16384 --tensor-parallel-size 8Size the nodes from the table above
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Rent 7× H200 SXM · BF16
- Estimated hourly price
- ~$25.13/hr
- est., whole server
- GPUs
- 7 GPUs
- 987 GB of VRAM
How Much VRAM Does Trinity Large Thinking Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 797.8 GB | 801.6 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Runs on 7× H200 SXM 141 GB |
Selected BF16
Brain floating point 16799.6GBat 16K context
- Weights
- 797.3 GB
- KV cache
- 2.0 GB
- Overhead
- 0.3 GB
Server-class: ~799.6 GB at 16K context · about 7× H200 SXM 141 GB
Which Hardware Can Run Trinity Large Thinking?
BF16 · 797.8 GBTrinity Large Thinking (BF16) requires 797.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1038+ GB is recommended. Using the full 262K context window can add up to 32.0 GB, bringing total usage to 829.8 GB. No consumer GPU has enough memory.
Needs server-class hardware
Trinity Large Thinking runs on a multi-GPU server · BF16 needs ~799.6 GB
| Hardware | Quant | Memory | Context | Hardware page | |
|---|---|---|---|---|---|
| 7× H200 SXM 141 GB · 987 GB | BF16 | 799.6 GB | 256K ctx | H200 SXM → | BF16 · 799.6 GB · 256K ctx |
| 5× B200 192 GB · 960 GB | BF16 | 799.6 GB | 256K ctx | B200 → | BF16 · 799.6 GB · 256K ctx |
By VRAM tier · typical GPUs and Macs at BF16
Consumer GPUs (up to 32 GB) and Macs up to 512 GB: too small for this model (needs ≥ 799.6 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
How good is Trinity Large Thinking?
llmrun Score blends public benchmark boards into one number; 100 = the average open model.
Not enough data yet for an llmrun Score.
A Score needs results on enough independent boards. How the Composite Score is built
Benchmarks
Here are all 5 boards that have results for this model, sorted by their weight in the llmrun Score; badges show who ran the test.
- LMArena Text
LMArena · Oct 2026Arena
Arena1367.3
#65/187 open
- LMArena WebDev
LMArena · Oct 2026Arena
Arena1237.9
#49/52 open
- SciCode
Epoch AI · Apr 2026Lab-reported
Lab-reported36.1%
#41/66 open
- Surface Evolver Bench
Epoch AI · Apr 2026Lab-reported
Lab-reported15.6%
#12/13 open
- CritPt
Epoch AI · Apr 2026Lab-reported
Lab-reported0.9%
#35/66 open
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.5 397B A17BAlibaba | 113.0 | 242.6 GB | —needs > 24 GB | same size |
| MiniMax M3MiniMax | 112.5 | 258 GB | —needs > 24 GB | same size |
| Llama 4 Maverick 17B 128E InstructMeta | 87.1 | 272.7 GB | —needs > 24 GB | same size |
| Llama 3.1 405B InstructMeta | 83.2 | 275.6 GB | —needs > 24 GB | same size |
| Intern S2 397BInternLM | — | 242.6 GB | —needs > 24 GB | same size |
| Ornith 1.5 397Bornith-ai | — | 242.6 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- Arcee AI
- Parameters
- 398.6B
- Active parameters
- 13.4B per token
- Architecture
- AfmoeForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 200,192
- Release Date
- 2026-04-01
- License
- Other
Frequently Asked Questions
How much VRAM does Trinity Large Thinking need?
Trinity Large Thinking requires 797.8 GB of VRAM at BF16. Full 262K context adds up to 32.0 GB (829.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 398.6B × 16 bits ÷ 8 = 797.3 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 32.5 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Trinity Large Thinking?
No — Trinity Large Thinking requires at least 797.8 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
Can I run Trinity Large Thinking on a Mac?
Trinity Large Thinking requires at least 797.8 GB at BF16. None of the 146 Macs we list has that much usable memory (the largest has 508.5 GB after macOS overhead).
Can I run Trinity Large Thinking locally?
Yes — Trinity Large Thinking can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
What's the download size of Trinity Large Thinking?
At BF16, the download is about 797.27 GB.
Which GPUs can run Trinity Large Thinking?
No single consumer GPU has enough VRAM to run Trinity Large Thinking at BF16 (797.8 GB). Multi-GPU or professional hardware is required.
Which devices can run Trinity Large Thinking?
Trinity Large Thinking requires at least 797.8 GB at BF16, which exceeds the usable memory of every unified-memory device we list (the largest has 508.5 GB).