whatmodel.sh / SOLAR 10.7B Instruct v1.0
SOLAR 10.7B Instruct v1.0
↗ share
+ contribute
SOLAR 10.7B Instruct v1.0 is an open-weights language model from Upstage, released December 2023. Its published size is 10.7B parameters and a 4k-token context window. Licence: cc-by-nc-4.0.
Catalogue identity notes
Arena entry 'solar-10.7b-instruct-v1.0' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'solar-10.7b-instruct-v1.0' as 'SOLAR-10.7B-Instruct-v1.0' (Upstage AI, CC-BY-NC-4.0), linking https://huggingface.co/upstage/SOLAR-10.7B-Instruct-v1.0. Maker repo upstage/SOLAR-10.7B-Instruct-v1.0 @090d405: licence tag cc-by-nc-4.0. 10,731,524,096 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2023-12: month the maker's Hugging Face repo was created (createdAt 2023-12-12).
$ how it compares
The LLMs ranked either side of SOLAR 10.7B Instruct v1.0, out of 828 with a score, then more from Upstage.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 660 |  Qwen3.5 0.8B | 38±1 | local | ||
| 661 |  Code Tulu 2 7B | 38±2 | local | ||
| 662 | SOLAR 10.7B Instruct v1.0 | 37±1 | local | ||
| 663 |  Phi-3.5-vision-instruct | 37±2 | local | ||
| 664 | Tulu 2 DPO 13B | 37±1 | local | ||
| more from Upstage | |||||
| 75 | Solar Open 2 (250B-A15B) | 72±1 | local | ||
| 170 | Solar Pro 4 | 65±2 | api $0.53/M | ||
| 240 | Solar Open 100B (102B-A12B) | 60±1 | local | ||
$ what it's good at
coding▁▁▁▃▆██▇▆▄▃▁ #400 of 458 · 76% of its score · 3 tests
Beats DeepSeek R1 Distill Qwen 7B, Qwen3.5 2B, DeepSeek R1 Distill Llama 8B here, though they rank higher overall. Trails Qwen2.5 1.5B Instruct, WizardLM 13B V1.2, Code Llama 7B Instruct here, though they rank lower overall.
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #349 of 393 * · 8.6% of its score · 2 tests
Beats Gemma 4 E2B, Code Llama 70B Instruct, Phi-3-mini-4k-instruct (June 2024 update) here, though they rank higher overall. Trails Gemma 3 270M IT, SmolLM2 360M Instruct, Granite 4.0 H 350M here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024) | 1064.0 ✎ | ▂▁▂▁▃▇█▆▇▄▂▂ beats 31/95 | 4.3% |
| LMArena (text) | 1152.4 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 42/379 | 4.3% |
other tests8 not in a task · 16% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| MedQA (USMLE, 4 options) | 54.7% ✎ | ▁▁··▄▁▁··▄▃█ beats 13/17 | 5.0% |
| CPTUB average (SpeakLeash, 0-5 judge scale) | 57.6% ✎ | ▁▂▂▂▃█▃▆▅▄▂▃ beats 24/95 | 3.2% |
| MedMCQA | 47.4% ✎ | ▂▂··▄▄▂▂·▄██ beats 10/17 | 2.7% |
| Polish EQ-Bench v2 (SpeakLeash) | 57.4% ✎ | ▁··▁▁▂▂▆▅▆█▆ beats 35/70 | 2.4% |
| Open PT-LLM Leaderboard average (9 Portuguese tasks) | 71.4% ✎ | ▃▁▁▁▁▁▁▁▂█▅▁ beats 184/276 | 0.9% |
| TriviaQA | 42.0% ✎ | ▁··▁█▃▃▁▄▅▄█ beats 17/28 | 0.8% |
| TruthfulQA MC2 | 71.7% ✎ | ▅▃█▆▁▂▅···▁▃ beats 31/31 | 0.4% |
| PubMedQA | 52.6% ✎ | ▃·▂▆·▄▂▄▄▂█▂ beats 0/17 | 0.3% |
$ 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 SOLAR 10.7B Instruct v1.0 · needs ~6.5 GB · from huggingface.co/mradermacherchat in the terminal
llama cli -hf mradermacher/SOLAR-10.7B-Instruct-v1.0-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf mradermacher/SOLAR-10.7B-Instruct-v1.0-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run SOLAR 10.7B Instruct v1.0 · needs ~6.5 GB · from huggingface.co/mradermacher
- Or download it in the terminal
lms get https://huggingface.co/mradermacher/SOLAR-10.7B-Instruct-v1.0-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run SOLAR 10.7B Instruct v1.0 · needs ~6.92 GB · from ollama.comchat in the terminal
ollama run solar:10.7b-instruct-v1-q4_K_M
Official weights and documentation: huggingface.co/upstage ↗
$ how the score was worked out · EvalPlus / EvalPerf published… 76%, Open Medical-LLM Leaderboard… 8%, Polyrating: A Cost-Effective… 4% and 5 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 |
|---|---|---|
 EvalPlus / EvalPerf published evaluation runs | 3 | 76% |
| Open Medical-LLM Leaderboard results (HF dataset openlifescienceai/results) | 3 | 8% |
 Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation (arXiv:2409.00696v3) | 1 | 4% |
 LMArena leaderboard dataset | 1 | 4% |
| SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark) | 1 | 3% |
| SpeakLeash Polish EQ-Bench leaderboard | 1 | 2% |
| Hallucinations Leaderboard results (HF dataset hallucinations-leaderboard/results) | 2 | 1% |
| Open Portuguese LLM Leaderboard (archived) | 1 | 1% |
| prior · pull toward the middle | 0% | |
| weighted average | = 37.4 |
$ 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