The story so far
Weights, every dated article, and repo commits on one rail.
19 repo commits since the weights landed
Teal is the weights (2023-12-12), grey ticks are repo commits.
SOLAR-10.7B-v1.0
upstage Apache 2.0 out 3 years ago
10.7B parameters, text in, text out, 4K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #625 in LLMs · score 73.5 method v1 Sources 3 Downloads 30d 13.4k · #484 of 619 tracked
Do people use it?13.4kdownloads, 30 days#484 of 619 we track · below the median+0 likes, downloads −1% this week324 likesHugging Face2026-10-24
Where to run it
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | Featherless | — | — | — | — | Oct 25 |
Where it's used
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 1 API provider; 21 quantized builds for local use; a big base for builders (60 fine-tunes, adapters and merges).
- As a service1Featherless
- On own hardware21quantized builds
- Built on6041 fine-tunes · 9 adapters · 10 merges
- In apps10Hugging Face Spaces
Top Spaces running it: mayafree/Model-Genome-Korea ♥ 33 · Yeyito/llm_contamination_detector ♥ 27 · CATIE-AQ/Guide_Evaluation_LLM ♥ 5 · officialhimanshu595/llama-factory ♥ 4
Will it fit?
Weights only, an estimate: leave room for the KV cache. Method
Fits a 16 GB card at INT4
About 6.4 GB at INT4, 13 GB at FP8, 26 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
What people built on it
81 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
81 derivatives on the Hub, most of them fine-tunes
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Reference
Everything else, folded: the full tables, every source.
Benchmarks
2 rows
Benchmarks 2 rows
measured third-party leaderboard rowauthor-reported number from the model card or eval files (2)No confidence intervals unless the source gave them; settings differ between rows, so compare within a benchmark, not across. Date = the benchmark's own date when it gives one, else when we read it.
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| author-reported H6 | 74.2 | SOLAR-10.7B-Instruct-v1.0 | model author | 2026-10-07 |
| author-reported H6 | 66.04 | SOLAR-10.7B-v1.0 | model author | 2026-10-07 |
What the card says
1 cited sentence
Summary generated, cited
The model uses a depth up-scaling methodology that integrates Mistral 7B weights into upscaled layers.[1]
Generated 2026-10-07 15:13 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
Use-case evidence
1 use case
What it is good at 1 use case
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Facts
20 fields
Listed in LLMs
Facts
- SOLAR-10.7B-v1.0 has 10.7B parameters. (safetensors metadata, observed 2026-10-24)
- Context length is 4,096 tokens (4K). (config.json max_position_embeddings, observed 2026-10-24)
- License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
- Created 2023-12-12, last modified 2024-09-10. (HF API, observed 2026-10-24)
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | LlamaForCausalLM |
| Model type | llama |
| Dtype | float16 |
| Params by dtype | F16: 10,731,524,096 |
| Size band | L |
| MoE | no |
| Context length | 4,096 (config.json max_position_embeddings) |
| Papers | arXiv:2312.15166 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
Servability
vLLM · SGLang
Servability
vLLM: supported · LlamaForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · LlamaForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Hardware fit
3 estimates
Hardware fit ESTIMATE
Weights-only memory for SOLAR-10.7B-v1.0: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 26 GB · Single high-end GPU (<80 GB) |
|---|---|
| FP8 (1 byte/param) | 13 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 6.4 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
All variants
40 variants
Lineage
finetune: 41quantized: 21merge: 10adapter: 9
Top variants (40 shown, ranked by downloads)
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| yanolja/YanoljaNEXT-EEVE-10.8B | finetune | — | 10.8B | L | apache-2.0 | yanolja | 92 | 526.3k |
| NousResearch/Nous-Hermes-2-SOLAR-10.7B | finetune | — | 10.7B | L | apache-2.0 | NousResearch | 210 | 421k |
| nlpai-lab/KULLM3 | finetune | — | 10.7B | L | apache-2.0 | nlpai-lab | 49 | 85.3k |
| Featherless-Chat-Models/SOLAR-10.7B-Instruct-v1.0 | finetune | — | 10.7B | L | cc-by-nc-4.0 | Featherless-Chat-Models | 0 | 67.7k |
| yanolja/KoSOLAR-10.7B-v0.2 | finetune | — | 10.8B | L | apache-2.0 | yanolja | 34 | 50.1k |
| jieunhan/TEST_MODEL | finetune | — | 19.3B | L | apache-2.0 | jieunhan | 0 | 33.5k |
| tanitan2/SOLAR-10.7B-Instruct-STOCK_SOLAR-ties | finetune | — | 10.7B | L | cc-by-nc-4.0 | tanitan2 | 0 | 30.2k |
| beomi/Solar-Ko-Recovery-11B | finetune | — | 11.0B | L | apache-2.0 | beomi | 24 | 24.5k |
| TheBloke/SOLAR-10.7B-v1.0-GGUF | quantized | GGUF | — | — | apache-2.0 | TheBloke | 14 | 20.3k |
| CallComply/SOLAR-10.7B-Instruct-v1.0-128k | finetune | — | 10.7B | L | cc-by-nc-4.0 | CallComply | 3 | 15.1k |
| ihopper/I-SOLAR-10.7B-dpo-sft-v1.0 | finetune | — | 10.7B | L | apache-2.0 | ihopper | 1 | 14.6k |
| ihopper/I-SOLAR-10.7B-sft-v1.0 | finetune | — | 10.7B | L | apache-2.0 | ihopper | 0 | 12.9k |
| mradermacher/SOLAR-10.7B-v1.0-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 11.5k |
| QuantFactory/SOLAR-10.7B-Instruct-v1.0-GGUF | quantized | GGUF | — | — | cc-by-nc-4.0 | QuantFactory | 1 | 10.4k |
| dustydecapod/Jovian-10.7B-v1.0 | finetune | — | 10.7B | L | apache-2.0 | dustydecapod | 1 | 8.9k |
| mradermacher/SOLAR-10.7B-v1.0-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 8.7k |
| chlee10/T3Q-platypus-SOLAR-10.7B-v1.0 | finetune | — | — | — | apache-2.0 | chlee10 | 0 | 8.3k |
| chlee10/T3Q-Platypus-SOLAR | finetune | — | — | — | apache-2.0 | chlee10 | 0 | 8.2k |
| QuantFactory/KULLM3-GGUF | quantized | GGUF | — | — | apache-2.0 | QuantFactory | 2 | 5.3k |
| MarsupialAI/Aqueducts-18B_iMatrix_GGUF | quantized | iMatrix | — | — | cc-by-nc-4.0 | MarsupialAI | 3 | 5.2k |
| QuantFactory/KoSOLAR-10.7B-v0.2-GGUF | quantized | GGUF | — | — | apache-2.0 | QuantFactory | 1 | 4.6k |
| srs6901/GGUF-SOLARized-GraniStral-14B_2102_YeAM-HCT_32QKV | merge | GGUF | — | — | apache-2.0 | srs6901 | 0 | 3.9k |
| DimensionSTP/Solar-Ko-Recovery-11B-scientific-qa | finetune | — | 10.7B | L | apache-2.0 | DimensionSTP | 0 | 2.8k |
| tensorblock/SOLAR-10.7B-v1.0-GGUF | quantized | GGUF | — | — | apache-2.0 | tensorblock | 0 | 2.2k |
| srs6901/GGUF-SOLARized-GraniStral-14B_2202_YeAM-HCT_X45QKV | merge | GGUF | — | — | apache-2.0 | srs6901 | 0 | 2.2k |
| featherless-ai-quants/upstage-SOLAR-10.7B-v1.0-GGUF | quantized | GGUF | — | — | — | featherless-ai-quants | 0 | 1.4k |
| srs6901/SOLARized-GraniStral-14B_2102_YeAM-HCT_32QKV | merge | GGUF | 13.9B | L | apache-2.0 | srs6901 | 0 | 1k |
| TheBloke/SOLAR-10.7B-v1.0-GPTQ | quantized | GPTQ | 10.7B | L | apache-2.0 | TheBloke | 2 | 831 |
| TheBloke/SOLAR-10.7B-v1.0-AWQ | quantized | AWQ | 10.7B | L | apache-2.0 | TheBloke | 1 | 737 |
| mlx-community/Nous-Hermes-2-SOLAR-10.7B-mlx | finetune | mlx | — | — | apache-2.0 | mlx-community | 0 | 617 |
| RefalMachine/ruadapt_solar_10.7_part1 | finetune | — | 10.7B | L | — | RefalMachine | 0 | 594 |
| MarsupialAI/Aqueducts-18B | finetune | — | 17.7B | L | cc-by-nc-4.0 | MarsupialAI | 5 | 306 |
| v000000/SyntheticMoist-11B | merge | — | 10.7B | L | — | v000000 | 1 | 477 |
| srs6901/SOLARized-GraniStral-14B_1902_YeAM-HCT | merge | GGUF | 13.9B | L | apache-2.0 | srs6901 | 0 | 494 |
| LoneStriker/Nous-Hermes-2-SOLAR-10.7B-6.0bpw-h6-exl2 | finetune | 0bpw | — | — | apache-2.0 | LoneStriker | 4 | 288 |
| mlx-community/SOLAR-10.7B-Instruct-v1.0-4bit | finetune | -4bit | — | — | cc-by-nc-4.0 | mlx-community | 0 | 453 |
| cgus/SOLAR-10.7B-Instruct-v1.0-128k-exl2 | finetune | exl2 | — | — | cc-by-nc-4.0 | cgus | 1 | 374 |
| LoneStriker/Nous-Hermes-2-SOLAR-10.7B-8.0bpw-h8-exl2 | finetune | 0bpw | — | — | apache-2.0 | LoneStriker | 3 | 269 |
| timpal0l/sol | finetune | — | 10.7B | L | apache-2.0 | timpal0l | 0 | 351 |
| fdisk/SOLAR-10.7B-Instruct-v1.0-Q4_K_M-GGUF | quantized | GGUF | — | — | cc-by-nc-4.0 | fdisk | 1 | 294 |
Repository changelog
19 commits
Repository changelog 19 commits
| Date | Commit | SHA |
|---|---|---|
| 2024-09-10 | Update README.md | 5a06298 |
| 2024-02-22 | Update the href URL link | a45090b |
| 2024-02-22 | Update README.md | 52e05c5 |
| 2023-12-28 | Update README.md | 399afd2 |
| 2023-12-27 | Update README.md | 2f4ae82 |
| 2023-12-26 | Update README.md | 1b9ad43 |
| 2023-12-25 | Update README.md | fbd7cd1 |
| 2023-12-25 | Upload solar_logo.png | 03434d7 |
| 2023-12-23 | Update README.md | d795fe1 |
| 2023-12-23 | Update README.md | c90fb13 |
| 2023-12-14 | Add license section | 9b879c7 |
| 2023-12-13 | Update README.md | 7fcda3e |
| 2023-12-13 | Update README.md | 04b884b |
| 2023-12-13 | Update README.md | 345cbf8 |
| 2023-12-12 | Update README.md | 6e27838 |
| 2023-12-12 | Update README.md | 15c954b |
| 2023-12-12 | Update README.md | 6853712 |
| 2023-12-12 | Add v1.0 files | a0667e9 |
| 2023-12-12 | initial commit | c78e13a |
All providers
1 provider · 1 endpoint
Where to run 1 providers · 1 endpoints
Hosted endpoints observed for SOLAR-10.7B-v1.0, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| Featherless hf | upstage/SOLAR-10.7B-v1.0 | — | — | — | — | — | — | — | live | 2026-10-25 |
Papers
1 paper
Papers 1
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Similar models
3 models
Similar models
- Llama-3.1-8B-Instruct · 8.03B same task (text-generation), more providers
- gemma-2-9b-it · 9.24B same size band (L), same task (text-generation), more providers
- Qwen3-8B · 8.19B same task (text-generation), more providers, permissive license
Sources
3 sources
Sources
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:41 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:13 UTC · Last verified 2026-10-25
Follow SOLAR-10.7B-v1.0
Watch this model
- /feeds/models/upstage--SOLAR-10.7B-v1.0.xml(RSS 2.0) · every change to SOLAR-10.7B-v1.0
- /feeds/families/upstage--SOLAR-v1.0.xml(RSS 2.0) · every change across the SOLAR-v1.0 family
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.