grok-3-beta: benchmark scores, price and ranking
grok-3-beta from xAI ranks #216 of 743 LLMs on the LLMs Tiger aggregate benchmark score (50.0), based on 5 benchmark results captured up to 2026-10-13.
- Organisation
- xAI
- Type
- Preview release
- Open weights
- No
- Aggregate rank
- #216 of 743 (score 50.0)
Relative to other models, grok-3-beta is strongest on Aider Polyglot (#18 of 49) and weakest on Fiction.liveBench (#28 of 47).
Across the 5 benchmarks where at least ten models are scored, grok-3-beta is in the top 10% on 0 and in the top half on 3.
grok-3-beta benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Aider Polyglot | 53.3% | #18 of 49 | Aider polyglot leaderboard | 2026-10-13 |
| BALROG | 29.5% | #15 of 34 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 75.8% | #126 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 55.6% | #100 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Fiction.liveBench | 58.3% | #28 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
About grok-3-beta
grok-3-beta is a preview release from xAI. No API price is listed for it yet. We hold 5 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where grok-3-beta stands
On our aggregate score, grok-3-beta is #216 of 743 ranked models, around the middle of the table. Its score is 50.0; the scale is centred on 50, so it sits 0.0 points above the typical model. That beats 71% of the ranked field. Among the 26 ranked xAI models it is #9. The next one up is grok-3-mini-beta (high) at 50.5. The next one down is grok-4-20 at 50.0.
grok-3-beta, benchmark by benchmark
Aider Polyglot puts grok-3-beta at 53.3%, #18 of 49. That beats 63% of the models tested, but 29.8 points short of gemini-2.5-pro-preview-06-05 (83.1%). The results either side are Quasar Alpha (54.7%) and Optimus Alpha (52.9%). It covers edits across 6 languages (225 exercises). The board median is 44.9%, so grok-3-beta is 8.4 points above it. Source: Aider polyglot leaderboard, captured 2026-10-13.
On BALROG, grok-3-beta reaches 29.5%. It ranks #15 of 34, ahead of 56% of the field, and 38.8 points behind GPT-6 Astra (max) at 68.3%. The results either side are claude-haiku-4-5-20251001_1K (31.2%) and Reka Flash 3 (29.2%). It covers long-horizon game-playing agents. The board median is 27.6%, so grok-3-beta is 1.9 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
grok-3-beta manages 75.8% on GPQA Diamond, good for #126 of 275. GPT-6 Astra (max) leads at 95.8%, 20.0 points further up, and 54% of models sit below. The results either side are gpt-oss-120b (high) (75.8%) and NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 (75.6%). It covers expert-level science questions. The board median is 71.7%, so grok-3-beta is 4.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
grok-3-beta lands below the middle on OTIS Mock AIME: 55.6%, #100 of 189, with 52% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is DeepSeek-R1-Distill-Qwen-32B at 55.6%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so grok-3-beta is 2.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
grok-3-beta lands below the middle on Fiction.liveBench: 58.3%, #28 of 47, with 57% of the field scoring higher. grok-4-0709 leads at 94.4%; the closest result above is GLM-4.5-FP8 at 58.3%. It covers long-context story comprehension (16k). The board median is 61.1%, so grok-3-beta is 2.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: solid, ahead of 51% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#126 of 275) and its weakest is OTIS Mock AIME (#100 of 189).
Reading these numbers
The 5 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (4), Aider polyglot leaderboard (1). 1 of them come from an official leaderboard and 4 from datasets or third-party runs. With 5 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #213 Qwen2.5-72B (50.1)
- #214 Gemini 3 Deep Think (Preview) ² (50.1)
- #215 glm-4.7-flash (50.1)
- #217 Claude Sonnet 4.6 (max) (50.0)
- #218 grok-4-20 (50.0)
- #219 step-5-preview-high (50.0)
More from xAI
- Grok 4.6 (xhigh) (#34, 53.6)
- grok-4-0709 (#38, 53.2)
- Grok 4 (#39, 53.2)
- Grok 4.7 (xhigh) (#41, 53.1)
- Grok 4.5 (high) (#85, 52.0)
- Grok 4.20 (Reasoning) (#136, 51.2)
Compare it on the live leaderboard → · How the aggregate score works