[whatmodel.sh](../../README.md) / InternLM2 Chat 20B
# InternLM2 Chat 20B
Shanghai AI Laboratory (InternLM) · released 2024-01 · 19.86B params · other
**score 42.5** ±1.3 · #527 of 732 · from 5 benchmark results across 5 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks InternLM2 Chat 20B among every model on that task or test, worst to best.
task vs the others rank counts tests
chat & writing ▁▁▁▃▆●▆█▆▇▇▃ #317 of 402* 71% 2
chat & writing · #317 of 402 · 71% of its score
Beats [Gemma 4 E4B](gemma-4-e4b.md), [OpenThinker3 7B](openthinker3-7b.md), [Code Llama 70B Instruct](codellama-70b-instruct.md) here, though they rank higher overall. Trails [LFM2 350M](lfm2-350m.md), [Gemma 3 1B](gemma-3-1b.md), [Qwen3 0.6B](qwen3-0-6b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AlpacaEval 2 length-controlled 18.8% ▁▂▂▃▆●█▆▄▄▁▁ beats 49/122 medium · avg 35% 45% sharp
Arena-Hard (v0.1, GPT-4-1106 judg… 24.4% ▂▁▁▅●▅▄▃▃▅▃▁ beats 23/60 medium · avg 47% 26% sharp
other tests (3)
test result vs the others in its run difficulty counts why
CPTUB average (SpeakLeash, 0-5 ju… 43.0% ▁▂●▂▃█▃▆▅▄▂▃ beats 10/95 medium · avg 65% 14% at its level
Polish EQ-Bench v2 (SpeakLeash) 36.5% ▁··▁▁▂●▆▅▆█▆ beats 10/70 medium · avg 62% 10% at its level
Open PT-LLM Leaderboard average (… 70.6% ▃▁▁▁▁▁▁▁▂●▅▁ beats 169/276 medium · avg 72% 4.0% at its level · noisy
source results share of the score
AlpacaEval 1.0 and 2.0 leaderboards 1 45% ▇▇▇▇▇
Arena-Hard-Auto v0.1 leaderboard 1 26% ▇▇▇
SpeakLeash CPTUB 1 14% ▇▇
SpeakLeash Polish EQ-Bench leaderboard 1 10% ▇
Open Portuguese LLM Leaderboard 1 4% ▇
prior: pull toward the middle 1% ▇
weighted average = 42.5
Licences: AlpacaEval 1.0 and 2.0 leaderboards ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · Arena-Hard-Auto v0.1 leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-v0-1-Apache-2.0.txt)) · SpeakLeash CPTUB ([MIT](../../licenses/hf-leaderboard-spaces-MIT.txt)) · SpeakLeash Polish EQ-Bench leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run InternLM2 Chat 20B (Q4_K_M, needs ~12 GB, from [huggingface.co/mradermacher](https://huggingface.co/mradermacher/internlm2-chat-20b-GGUF/tree/43825e5e0b18dfd35c413d7a41f1d6ebcc38041a))
```sh
llama cli -hf mradermacher/internlm2-chat-20b-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf mradermacher/internlm2-chat-20b-GGUF:Q4_K_M
```
Other versions: Q8_0 (~21.1 GB) `llama cli -hf mradermacher/internlm2-chat-20b-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run InternLM2 Chat 20B (Q4_K_M, needs ~12 GB, from [huggingface.co/mradermacher](https://huggingface.co/mradermacher/internlm2-chat-20b-GGUF/tree/43825e5e0b18dfd35c413d7a41f1d6ebcc38041a))
```sh
lms get https://huggingface.co/mradermacher/internlm2-chat-20b-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~21.1 GB) `lms get https://huggingface.co/mradermacher/internlm2-chat-20b-GGUF@Q8_0 -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run InternLM2 Chat 20B (Q4_K_M, needs ~12 GB, from [huggingface.co/mradermacher](https://huggingface.co/mradermacher/internlm2-chat-20b-GGUF/tree/43825e5e0b18dfd35c413d7a41f1d6ebcc38041a))
```sh
ollama run hf.co/mradermacher/internlm2-chat-20b-GGUF:Q4_K_M
```
Other versions: Q8_0 (~21.1 GB) `ollama run hf.co/mradermacher/internlm2-chat-20b-GGUF:Q8_0`
### Other ways to run it
1. Run it as published (BF16, needs ~39.72 GB, from [huggingface.co/internlm](https://huggingface.co/internlm/internlm2-chat-20b/tree/d606d418463ca058a8338e6ce4ba0f200aab15c1))
```sh
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='internlm/internlm2-chat-20b', torch_dtype='auto', device_map='auto', trust_remote_code=True)('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
```
Model repository and documentation: [huggingface.co/internlm](https://huggingface.co/internlm/internlm2-chat-20b)
[interactive version](https://whatmodel.sh/model/internlm2-chat-20b) · [JSON](https://whatmodel.sh/api?format=json&all=1)