whatmodel.sh / DeepSeek-TNG R1T2 Chimera

DeepSeek-TNG R1T2 Chimera

+ contribute

DeepSeek-TNG R1T2 Chimera is an open-weights language model from TNG Technology Consulting, released July 2025. Its published size is 684.53B parameters (37B active per token). Licence: mit.

Catalogue identity notes

Maker repo tngtech/DeepSeek-TNG-R1T2-Chimera @748c3e0: licence tag mit. 684,531,386,000 parameters from the Hugging Face safetensors metadata; active 37B per token (DeepSeek-V3 architecture (DeepseekV3ForCausalLM, 671B total); DeepSeek-V3 report: 37B activated per token). Released 2025-07: month the maker's Hugging Face repo was created (createdAt 2025-07-02).

51.1±10.2#413 of 949 LLMsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of DeepSeek-TNG R1T2 Chimera, out of 1132 with a score.

#	model	score	where	RAM	released
539	xLAM-2-1b-fc-r Salesforce	51±4	local	~2.6GB	2025-03
540	Hammer2.1-3b MadeAgents	51±4	local	~3.3GB	2024-12
541	DeepSeek-TNG R1T2 Chimera TNG Technology Consulting	51±10	local	~229GB	2025-07
542	Llama 3.1 Nemotron 51B Instruct NVIDIA	51±2	local	~34GB	2024-09
543	OpenHands LM 32B v0.1 All Hands AI	51±3	local	~21GB	2025-03

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests1 not in a task · 54% of its score

test	result	vs the others in its run	difficulty	counts	why
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)	94.5% ✎	▁▁▂▃▄▄█▆▃▂▂▁ beats 53/140	very easy · avg 94%	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 54%	blunt · maxed out · noisy

$ where to run it

llama.cpp · recommended

version

Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
Windows
irm https://llama.app/install.ps1 | iex
other ways to install ↗
Run DeepSeek-TNG R1T2 Chimera · needs ~404.9 GB · from huggingface.co/unsloth
chat in the terminal
llama cli -hf unsloth/DeepSeek-TNG-R1T2-Chimera-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf unsloth/DeepSeek-TNG-R1T2-Chimera-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run DeepSeek-TNG R1T2 Chimera · needs ~404.9 GB · from huggingface.co/unsloth
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/unsloth/DeepSeek-TNG-R1T2-Chimera-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Other ways to run it

Run it as published · needs ~688.59 GB · from huggingface.co/tngtech (official)
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='tngtech/DeepSeek-TNG-R1T2-Chimera', torch_dtype='auto', device_map='auto', trust_remote_code=True)('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/tngtech ↗

$ how the score was worked out · Vectara hallucination leaderb… 54%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

source	results	share of the score	licence
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)	1	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 54%	Apache-2.0
prior · pull toward the middle		▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 46%
weighted average		= 51.1

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated 2026-10-11 02:20 UTC · sources and licences · json