LLMs · Ai2
TMax 4B
- Parameters
- 4.2B
- Architecture
- Dense
- Max context
- 256K tokens
- License
- apache-2.0
- Released
- 2026-06
How to run it
Original weights · Hugging Face
Full-precision files from Ai2, for fine-tuning, vLLM, Transformers.
Memory needed
Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.
| Quantization | Quality | Weights | Total at 8K context | Total at 32K context | Download |
|---|---|---|---|---|---|
Q2_K |
Low | 1.7 GB | 2.6 GB | 3.4 GB | All files (opens in a new tab) |
Q3_K_M |
Low | 2.1 GB | 2.9 GB | 3.7 GB | All files (opens in a new tab) |
Q4_K_M |
Good | 2.6 GB | 3.4 GB | 4.2 GB | All files (opens in a new tab) |
Q5_K_M |
High | 3.0 GB | 3.9 GB | 4.7 GB | All files (opens in a new tab) |
Q6_K |
High | 3.5 GB | 4.3 GB | 5.1 GB | All files (opens in a new tab) |
Q8_0 |
Very high | 4.5 GB | 5.3 GB | 6.1 GB | All files (opens in a new tab) |
F16 |
Full | 8.4 GB | 9.3 GB | 10.1 GB | Original weights (opens in a new tab) |
Best hardware for TMax 4B
The least expensive GPU in our catalog that runs it without offloading is the Intel Arc A380 (launch price $139) at Q4_K_M.
| Grade · points | Hardware | Quant | Memory | Speed | Fit |
|---|---|---|---|---|---|
| 82 ptsGreat (A) · 82 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 4060 | Q8_0 | 5.7 GB / 8.0 GB | 40.4 tok/s | Runs well |
| 82 ptsGreat (A) · 82 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 4070 Ti SUPER | F16 | 10.1 GB / 16.0 GB | 46.0 tok/s | Runs well |
| 82 ptsGreat (A) · 82 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 4080 SUPER | F16 | 10.1 GB / 16.0 GB | 50.3 tok/s | Runs well |
| 82 ptsGreat (A) · 82 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 5070 Ti | F16 | 10.1 GB / 16.0 GB | 61.3 tok/s | Runs well |
| 82 ptsGreat (A) · 82 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 5080 | F16 | 10.1 GB / 16.0 GB | 65.7 tok/s | Runs well |
| 78 ptsGreat (A) · 78 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 11.4/15 | Intel Arc B580 | Q8_0 | 5.9 GB / 12.0 GB | 51.7 tok/s | Runs well |
| 78 ptsGreat (A) · 78 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 11.4/15 | NVIDIA GeForce RTX 3060 12GB | Q8_0 | 5.9 GB / 12.0 GB | 53.4 tok/s | Runs well |
| 78 ptsGreat (A) · 78 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 11.4/15 | NVIDIA GeForce RTX 4070 SUPER | Q8_0 | 5.9 GB / 12.0 GB | 74.8 tok/s | Runs well |
| 77 ptsGreat (A) · 77 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.1/15 | AMD Radeon RX 7900 XTX | F16 | 10.5 GB / 24.0 GB | 56.6 tok/s | Runs well |
| 77 ptsGreat (A) · 77 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.1/15 | NVIDIA GeForce RTX 3090 | F16 | 10.5 GB / 24.0 GB | 64.0 tok/s | Runs well |
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.