whatmodel.sh / DeepSeek-V2.5 (1210)
DeepSeek-V2.5 (1210)
↗ share
+ contribute
DeepSeek-V2.5 (1210) is an open-weights language model from DeepSeek, released December 2024. Its published size is 236B parameters (21B active per token). Licence: deepseek.
Catalogue identity notes
Comparison-only. Arena entry 'deepseek-v2.5-1210' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'deepseek-v2.5-1210' as 'Deepseek-v2.5-1210' (DeepSeek, DeepSeek), linking https://huggingface.co/deepseek-ai/DeepSeek-V2.5-1210. Maker repo deepseek-ai/DeepSeek-V2.5-1210 @6f134cb: licence tag other (deepseek). Parameters: DeepSeek-V2 architecture: 236B total / 21B active (HF safetensors metadata 235.7B). Released 2024-12: month the maker's Hugging Face repo was created (createdAt 2024-12-10).
$ how it compares
The LLMs ranked either side of DeepSeek-V2.5 (1210), out of 828 with a score, then more from DeepSeek.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 299 |  Granite 4.1 30B | 56±1 | local | ||
| 300 |  Mistral Small 3.2 24B | 56±1 | api $0.11/M +local | ||
| 301 | DeepSeek-V2.5 (1210) | 56±1 | local | ||
| 302 |  Qwen3 8B | 56±1 | api $0.31/M +local | ||
| 303 | DeepSeek R1 Distill Qwen 14B | 56±1 | local | ||
| more from DeepSeek | |||||
| 21 | DeepSeek V4.1 Flash | 78±1 | api $0.26/M +local | ||
| 31 | DeepSeek V4 Pro (0813) | 76±1 | api $1.63/M +local | ||
| 35 | DeepSeek V3.2 Speciale | 76±3 | comparison only | ||
$ what it's good at
coding▁▁▁▃▆██▇▆▄▃▁ #201 of 458 * · 65% of its score · 2 tests
Beats Qwen3.5 4B, Qwen3-VL 32B Instruct, Falcon-H1R 7B here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 3B Reasoning 2512, DeepSeek-Coder-V2 Instruct (0724) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Aider polyglot (225 exercises, two attempts) | 17.8% ✎ | ▂·▂▃▅▂▃█▆▄▂▂ beats 8/49 | 36% |
| Aider code editing (133 Exercism Python exercises, two attempts) | 72.2% ✎ | ▁·▁▂▄▁▆▄█▆▂▂ beats 53/62 | 30% |
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #192 of 393 * · 26% of its score · 2 tests
Beats Granite 4.2 8B, Qwen3-VL 30B-A3B Thinking, Mistral Small 4 (119B-A6.5B) here, though they rank higher overall. Trails LFM2.5 1.2B Instruct, LFM2.5 1.2B Thinking, Gemma 3 12B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena (text) | 1323.4 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 159/379 | 18% |
| IFEval prompt-level strict | 79.3% ✎ | ▄▄··▄·▄··█·▄ beats 1/6 | 8.5% |
math▁▂▃▅▆▅▄█▄▁▁▁ #225 of 398 * · 6.9% of its score · 1 test
Beats GPT-4o (2024-08-06), Command A (03-2025), GPT-4o (2024-11-20) here, though they rank higher overall. Trails Olmo 3 7B Instruct, DeepSeek R1 Distill Llama 8B, Phi-4-mini-reasoning 3.8B here, though they rank lower overall.
other tests1 not in a task · 0.9% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| DROP F1, 3-shot (simple-evals) | 89.1% ✎ | ▅▂▆█▆·▃·▃▅·▂ beats 19/23 | 0.9% |
$ 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 | shWindowsirm https://llama.app/install.ps1 | iex - Run DeepSeek-V2.5 (1210) · needs ~142.45 GB · from huggingface.co/bartowskichat in the terminal
llama cli -hf bartowski/DeepSeek-V2.5-1210-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf bartowski/DeepSeek-V2.5-1210-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run DeepSeek-V2.5 (1210) · needs ~142.45 GB · from huggingface.co/bartowski
- Or download it in the terminal
lms get https://huggingface.co/bartowski/DeepSeek-V2.5-1210-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
MLX (Mac)
version
- Install MLX (Mac) · Apple silicon Macs only, text only.with uv
uv tool install mlx-lmorpip install mlx-lm - Run DeepSeek-V2.5 (1210) · needs ~132.7 GB, Mac only · from huggingface.co/mlx-communitychat in the terminal
mlx_lm.chat --model mlx-community/DeepSeek-V2.5-1210-4bit - Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/DeepSeek-V2.5-1210-4bit
Weights and documentation: huggingface.co/deepseek-ai ↗
$ how the score was worked out · Aider polyglot benchmark runs 36%, Aider code editing benchmark… 30%, LMArena leaderboard dataset 18% and 2 more
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 |
|---|---|---|
 Aider polyglot benchmark runs | 1 | 36% |
 Aider code editing benchmark runs (old edit leaderboard) | 1 | 30% |
 LMArena leaderboard dataset | 1 | 18% |
| OpenCompass CompassAcademic leaderboard (full version) Space: committed predictions file | 2 | 9% |
| DeepSeek vendor-reported benchmark scores for its own models (Hugging Face model cards) · lab-reported | 1 | 7% |
| prior · pull toward the middle | 1% | |
| weighted average | = 55.8 |
$ improve this page
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 · sources and licences · json