RAG and retrieval
Results are only compared within a matching group, so a model tested one way is never ranked against a model tested another way.
Retrieve relevant context and ground an answer
Ranking groups use separate scales. Compare configurations only inside the same group; EvalRank never flattens unlike systems into one leaderboard.
Cross-benchmark consensus
One merged ranking of models by where they place across this topic's benchmark boards. Derived from the boards below, not a calibrated top set; it updates as sources are re-scraped.
| Rank | Model | Score | Benchmarks |
|---|---|---|---|
| 1 | gpt-6 sol facts-grounding-v2:gpt-6-sol | 0.962 | 1 |
| 2 | gpt-6 astra facts-grounding-v2:gpt-6-astra | 0.926 | 1 |
| 3 | gemma 4 26b a4b facts-grounding-v2:gemma-4.26b-a4b-it | 0.891 | 1 |
| 4 | gemma 4 31b facts-grounding-v2:gemma-4.31b-it | 0.857 | 1 |
| 5 | gpt-5.2 facts-grounding-v2:gpt-5.2 | 0.825 | 1 |
| 6 | gpt-5.5 facts-grounding-v2:gpt-5.5 | 0.793 | 1 |
| 7 | gemini 3.7 flash facts-grounding-v2:gemini-3.7-flash | 0.763 | 1 |
| 8 | grok 4.5 facts-grounding-v2:grok-4.5.0708 | 0.733 | 1 |
| 9 | gemini 2.5 pro facts-grounding-v2:gemini-2.5-pro | 0.705 | 1 |
| 10 | gemini 3.5 flash facts-grounding-v2:gemini-3.5-flash | 0.677 | 1 |
| 11 | gemini 3.8 flash facts-grounding-v2:gemini-3.8-flash | 0.651 | 1 |
| 12 | gemini 3.6 flash facts-grounding-v2:gemini-3.6-flash | 0.625 | 1 |
| 13 | llama 3 - grounded language model facts-grounding-v2:llama-3-glm-v2 | 0.600 | 1 |
| 14 | gpt-5.6 terra facts-grounding-v2:gpt-5.6-terra | 0.576 | 1 |
| 15 | gemini 3.5 flash-lite facts-grounding-v2:gemini-3.5-flash-lite | 0.552 | 1 |
| 16 | gpt-5.4 facts-grounding-v2:gpt-5.4 | 0.529 | 1 |
| 17 | gemini 2.5 flash facts-grounding-v2:gemini-2.5-flash | 0.507 | 1 |
| 18 | claude opus 4.8 facts-grounding-v2:claude-opus-4.8-default | 0.486 | 1 |
| 19 | gpt-5 facts-grounding-v2:gpt-5 | 0.465 | 1 |
| 20 | gpt-5.6 sol facts-grounding-v2:gpt-5.6-sol | 0.444 | 1 |
| 21 | claude opus 5.5 facts-grounding-v2:claude-opus-5.5-default | 0.425 | 1 |
| 22 | gemini 3.1 flash-lite preview facts-grounding-v2:gemini-3.1-flash-lite-preview | 0.405 | 1 |
| 23 | gemini 3.1 pro preview facts-grounding-v2:gemini-3.1-pro-preview | 0.387 | 1 |
| 24 | claude opus 4.5 facts-grounding-v2:claude-opus-4.5.20251101 | 0.368 | 1 |
| 25 | claude sonnet 4.5 (thinking) facts-grounding-v2:claude-sonnet-4.5-thinking-20250929 | 0.351 | 1 |
| 26 | claude opus 4.7 facts-grounding-v2:claude-opus-4.7-default | 0.333 | 1 |
| 27 | gemini 3 flash preview facts-grounding-v2:gemini-3-flash-preview | 0.316 | 1 |
| 28 | claude sonnet 4.5 facts-grounding-v2:claude-sonnet-4.5.20250929 | 0.300 | 1 |
| 29 | gemma 3 12b facts-grounding-v2:gemma-3.12b-it | 0.284 | 1 |
| 30 | gpt-5 mini facts-grounding-v2:gpt-5-mini | 0.268 | 1 |
| 31 | deepseek-v3.1 facts-grounding-v2:deepseek-v3.1 | 0.253 | 1 |
| 32 | gemma 3 27b facts-grounding-v2:gemma-3.27b-it | 0.238 | 1 |
| 33 | claude sonnet 4.6 facts-grounding-v2:claude-sonnet-4.6-default | 0.224 | 1 |
| 34 | claude sonnet 4 facts-grounding-v2:claude-sonnet-4.20250514 | 0.209 | 1 |
| 35 | deepseek-r1 facts-grounding-v2:deepseek-r1.0528 | 0.195 | 1 |
| 36 | claude opus 4.6 facts-grounding-v2:claude-opus-4.6-default | 0.182 | 1 |
| 37 | claude opus 4.1 facts-grounding-v2:claude-opus-4.1.20250805 | 0.169 | 1 |
| 38 | grok 4 facts-grounding-v2:grok-4.0709 | 0.156 | 1 |
| 39 | gpt-5.1 facts-grounding-v2:gpt-5.1 | 0.143 | 1 |
| 40 | mistral large 2 facts-grounding-v2:mistral-large-2411 | 0.130 | 1 |
| 41 | qwen 3 235b a22b thinking facts-grounding-v2:qwen3.235b-a22b-thinking-2507 | 0.118 | 1 |
| 42 | gpt-4.1 facts-grounding-v2:gpt-4.1 | 0.106 | 1 |
| 43 | grok 3 mini facts-grounding-v2:grok-3-mini | 0.095 | 1 |
| 44 | mistral small 3.1 facts-grounding-v2:mistral-small-2503 | 0.083 | 1 |
| 45 | grok-4-fast-reasoning facts-grounding-v2:grok-4-fast-reasoning | 0.072 | 1 |
| 46 | qwen 3 235b a22b instruct 2506 facts-grounding-v2:qwen3.235b-a22b-instruct-2507 | 0.061 | 1 |
| 47 | gemma 3 4b facts-grounding-v2:gemma-3.4b-it | 0.051 | 1 |
| 48 | o3 facts-grounding-v2:o3 | 0.040 | 1 |
| 49 | deepseek-v3 facts-grounding-v2:deepseek-v3 | 0.030 | 1 |
| 50 | gpt-oss-120b facts-grounding-v2:gpt-oss-120b | 0.020 | 1 |
| 51 | o4 mini facts-grounding-v2:o4-mini | 0.010 | 1 |
| 52 | gpt-oss-20b facts-grounding-v2:gpt-oss-20b | 0.000 | 1 |
Model configuration ยท Direct prompt
Exploratory - not a ranked winner yetrg-rag-retrieval-model-configuration-direct-prompt-model-configuration-v1Explorer evidence
FACTS Grounding v2
Single source. Observed ; fresh until .
| Rank | Configuration | Capability | Uncertainty | Families |
|---|---|---|---|---|
| 1 | GPT-6 Sol facts-grounding-v2:gpt-6-solconfig_8ed4d1395671d376bcfee335ffeb8d5f8038c7fe04da1be97943d4e889520160 | 85% | Unknown | 1 |
| 2 | GPT-6 Astra facts-grounding-v2:gpt-6-astraconfig_133c9e505fd788462ac768c4805bef5b5fe59e2324bb1f8fff6e845bcddb3872 | 82% | Unknown | 1 |
| 3 | Gemma 4 26B A4B facts-grounding-v2:gemma-4-26b-a4b-itconfig_55069a9aaccbd5ee73d7a243cd0c8df03b39b412016cf1a0b7b837edafff174d | 81% | Unknown | 1 |
| 4 | Gemma 4 31B facts-grounding-v2:gemma-4-31b-itconfig_fa8dd1d921924622402184bc1cc37f6f1c4a9ab5a8550e75c01250d3132198cf | 81% | Unknown | 1 |
| 5 | GPT-5.2 facts-grounding-v2:gpt-5.2-2025-12-11config_e0c03c8f9bae1f20e36791d9750407056b2aa6b1553412055e7d3c3ccca15500 | 76% | Unknown | 1 |
| 6 | GPT-5.5 facts-grounding-v2:gpt-5.5-2026-04-23config_710a55db402450b2e005fc3bafb7df1760041ff0ffd1cf075609e8ee18a32f44 | 76% | Unknown | 1 |
| 7 | Gemini 3.7 Flash facts-grounding-v2:gemini-3.7-flashconfig_4026e787f2d1ee38a44d0ba47d67fa9dfea76bcb5a79856e0fceed78e525a448 | 76% | Unknown | 1 |
| 8 | Grok 4.5 facts-grounding-v2:grok-4.5-0708config_4a62b1b27883f8102aa668987be5f844ca1c997adb4d5533520aa017cce749c3 | 75% | Unknown | 1 |
| 9 | Gemini 2.5 Pro facts-grounding-v2:gemini-2.5-proconfig_783f2ed58a967605d08ca2b3dcb6b87ebb23f40233b12a6fdea74b2da27c4ae3 | 74% | Unknown | 1 |
| 10 | Gemini 3.5 Flash facts-grounding-v2:gemini-3.5-flashconfig_4acc24e88a40e43ac0715989ac6c43eacc8d6c0725d94ad82e27d2d6d7a05742 | 73% | Unknown | 1 |
| 11 | Gemini 3.8 Flash facts-grounding-v2:gemini-3.8-flashconfig_289038458a43165e74dc5110c147d0ee5e4582c3c45ee8e794afb02b454cbea6 | 73% | Unknown | 1 |
| 12 | Gemini 3.6 Flash facts-grounding-v2:gemini-3.6-flashconfig_840da752a99e47721a415ce8da992542e782883bd3615b3dbd61a6e1f4a3de1c | 73% | Unknown | 1 |
| 13 | Llama 3 - Grounded Language Model facts-grounding-v2:llama-3-glm-v2config_0f06e4f0f5d94989ce492f699fcfcc6885ac22132e3f2ec0eee1af5989a66970 | 72% | Unknown | 1 |
| 14 | GPT-5.6 Terra facts-grounding-v2:gpt-5.6-terraconfig_c000ae6e53b83f053539736253672afb6731d79b61281f862e6b43a61ab41378 | 71% | Unknown | 1 |
| 15 | Gemini 3.5 Flash-Lite facts-grounding-v2:gemini-3.5-flash-liteconfig_9c30a1f488adc61f4ed665cb68266f3ce5897a9eda2c668ccba6b873c9adfdbd | 71% | Unknown | 1 |
| 16 | GPT-5.4 facts-grounding-v2:gpt-5.4-2026-03-05config_6418c3c2e203ed18c03b56f3ca6d134f627f1a5d43b19b0204e107b2f56261da | 70% | Unknown | 1 |
| 17 | Gemini 2.5 Flash facts-grounding-v2:gemini-2.5-flashconfig_1574f46274d945b78a6e511b08d21e94fb8b4a3ab6342870d3597f6508f1f54f | 70% | Unknown | 1 |
| 18 | Claude Opus 4.8 facts-grounding-v2:claude-opus-4-8-defaultconfig_70ceef0d4d2e5ded2813f951e152ae2c55d8049c6b41f29a1e8adb3c5107b9bf | 70% | Unknown | 1 |
| 19 | GPT-5 facts-grounding-v2:gpt-5-2025-08-07config_77323bbac142dd39ba7c4bf6b7d7767fcda3d4766d121d6cd6668b0ecc0d0c7f | 70% | Unknown | 1 |
| 20 | GPT-5.6 Sol facts-grounding-v2:gpt-5.6-solconfig_beb84de3aa4af33990738adbdf4adc429c91cbccad74093888a0e1c61e657be8 | 69% | Unknown | 1 |
| 21 | Claude Opus 5.5 facts-grounding-v2:claude-opus-5-5-defaultconfig_d2c7c107ab0570cb61aecf7976c1bf66a5f97bfb7edaa45f27a21886281c4c92 | 68% | Unknown | 1 |
| 22 | Gemini 3.1 Flash-Lite Preview facts-grounding-v2:gemini-3.1-flash-lite-previewconfig_99ace85b1a0db8ac88c5925d2501222a7cec3dee6c4f2c9fd7dc8910c6194a29 | 67% | Unknown | 1 |
| 23 | Gemini 3.1 Pro Preview facts-grounding-v2:gemini-3.1-pro-previewconfig_5e7d3db54d056747b7f5db367b48e115c38aff0a37a99c447039ea568c28e338 | 66% | Unknown | 1 |
| 24 | Claude Opus 4.5 facts-grounding-v2:claude-opus-4-5-20251101config_8e683bad552a0278e187fdecbb5ac3cc120da9e5446f8d82f8d6beed69b7f36e | 62% | Unknown | 1 |
| 25 | Claude Sonnet 4.5 (thinking) facts-grounding-v2:claude-sonnet-4-5-thinking-20250929config_1b4c3228265e7e479e73ed07f8f2892c3963888cc8563baabb061799a56faff0 | 62% | Unknown | 1 |
| 26 | Claude Opus 4.7 facts-grounding-v2:claude-opus-4-7-defaultconfig_2e631e3100b7beeb70513da3ace5ee7e57a63f8ed3251258ecdab19556959e11 | 62% | Unknown | 1 |
| 27 | Gemini 3 Flash Preview facts-grounding-v2:gemini-3-flash-previewconfig_20eb6b747fa901d453eaa5033783644b6c431e397161655e150531260876035f | 59% | Unknown | 1 |
| 28 | Claude Sonnet 4.5 facts-grounding-v2:claude-sonnet-4-5-20250929config_95e588303138355db334fa38eeb6f8cca3d2934a7719861f58044a1b6ab8132c | 59% | Unknown | 1 |
| 29 | Gemma 3 12B facts-grounding-v2:gemma-3-12b-itconfig_a9c455a70f834b2129dff49c1ece1f2927acb1ffd63b7659341d7a045437a997 | 58% | Unknown | 1 |
| 30 | GPT-5 mini facts-grounding-v2:gpt-5-mini-2025-08-07config_b2b1657763d97067baa04e89161538070afa3423ebc3247eaec84f3778ced376 | 58% | Unknown | 1 |
| 31 | deepseek-v3.1 facts-grounding-v2:deepseek-v3.1config_ee4508056235cade80f66a936ba0cdab3df381537dbf475dd68b4d7fc6e0a92d | 57% | Unknown | 1 |
| 32 | Gemma 3 27B facts-grounding-v2:gemma-3-27b-itconfig_e19ed0df0ca6248946f01dd2a42d65257e961a7df19cebd2ee47e354366d1825 | 56% | Unknown | 1 |
| 33 | Claude Sonnet 4.6 facts-grounding-v2:claude-sonnet-4-6-defaultconfig_a919001f9d86cbd5e49ee79cdfd735c95db3fd63b883e84698ad2e20d2ab387a | 56% | Unknown | 1 |
| 34 | Claude Sonnet 4 facts-grounding-v2:claude-sonnet-4-20250514config_9043e119a59e2193df648bb3c6afe51568f3c055f0bf19deee19d1d225ef4e0d | 56% | Unknown | 1 |
| 35 | DeepSeek-R1 facts-grounding-v2:deepseek-r1-0528config_cb79531fc777a670a4d1b7cbc9a640217d7214083963cfc8d24921c5100cb949 | 56% | Unknown | 1 |
| 36 | Claude Opus 4.6 facts-grounding-v2:claude-opus-4-6-defaultconfig_9ab41003aafa024c3cf8366a04bedf51180b6cfce8a4d16f225194a9ba8c5a3f | 55% | Unknown | 1 |
| 37 | Claude Opus 4.1 facts-grounding-v2:claude-opus-4-1-20250805config_4dc0fe5957c9d32e28017954cd60bd767aaf5faff46cab7c7036d0ded3698128 | 55% | Unknown | 1 |
| 38 | Grok 4 facts-grounding-v2:grok-4-0709config_bbd27cb84ac96d337e1e2d8df725f8fb0609c1629ff37871f886a6860bc6e201 | 55% | Unknown | 1 |
| 39 | GPT-5.1 facts-grounding-v2:gpt-5.1-2025-11-13config_e903e9c3e67d92ca7f2f2f52ec4d8f569d3b32adbd32d1cf3da84f3e3ea340d3 | 50% | Unknown | 1 |
| 40 | Mistral Large 2 facts-grounding-v2:mistral-large-2411config_d5cfa8378c3323b246494490408304aa1b6340c9cb0b499bf0eab4d96fb0089a | 50% | Unknown | 1 |
| 41 | Qwen 3 235B A22B Thinking facts-grounding-v2:qwen3-235b-a22b-thinking-2507config_ea118d5f5f30dbff56f980ab792e6c229143152fb88c3371ce3958d1e9c6b38e | 48% | Unknown | 1 |
| 42 | GPT-4.1 facts-grounding-v2:gpt-4.1-2025-04-14config_6955334d2ec8f0e6cdc8f3fdfd1ac1fae49568f75e36e7ce519de046167131b4 | 46% | Unknown | 1 |
| 43 | Grok 3 Mini facts-grounding-v2:grok-3-miniconfig_64ed484ee4d4578b51e027b889a16bfde637ab58ae70225a8106071a28b8c9f4 | 46% | Unknown | 1 |
| 44 | Mistral Small 3.1 facts-grounding-v2:mistral-small-2503config_b4d44ad7cfb5653b4324be8f112c20d794a938815ab9c05394da6a6ff5da6e61 | 43% | Unknown | 1 |
| 45 | grok-4-fast-reasoning facts-grounding-v2:grok-4-fast-reasoningconfig_350abae8c01cdbcb531f1024364a1fd54de8f68c8ffefd639a9e9a7c05a23ba4 | 43% | Unknown | 1 |
| 46 | Qwen 3 235B A22B Instruct 2506 facts-grounding-v2:qwen3-235b-a22b-instruct-2507config_f9942a332bf65fb4c20a6251bddb32d98f2a3fa5b2b996ea19eaa4aff463e301 | 41% | Unknown | 1 |
| 47 | Gemma 3 4B facts-grounding-v2:gemma-3-4b-itconfig_e1dc7a7cb1f75000356fc5d8b8ea6dac194982fb867aa45f0fe41c7c0e22c02b | 38% | Unknown | 1 |
| 48 | o3 facts-grounding-v2:o3-2025-04-16config_aa1102e1f28d441968eb0579c5022f260305a70f81f06e09c8761c0257f05c74 | 36% | Unknown | 1 |
| 49 | DeepSeek-V3 facts-grounding-v2:deepseek-v3config_b6fb1eb16d428ab7efa0b27b7005ed0621847787a9795d107fe4d9daf859fd35 | 33% | Unknown | 1 |
| 50 | gpt-oss-120b facts-grounding-v2:gpt-oss-120bconfig_a76962bdae4d7535750df3b65733135f2ed5f454f059573b12adbef178ba23ec | 31% | Unknown | 1 |
| 51 | o4 mini facts-grounding-v2:o4-mini-2025-04-16config_486c3d4718644118c12186a60b8248565fe8c255fd8023a8d979639fb0ab2db6 | 29% | Unknown | 1 |
| 52 | gpt-oss-20b facts-grounding-v2:gpt-oss-20bconfig_3795d71732182be0ae6265dba51dfda1f7fa89bcef6f838a60635ff3441f757a | 23% | Unknown | 1 |
Not enough independent evidence for a top set
- Calibration has not been validated
- Independent benchmark-family coverage is below the required threshold