Skip to content
AI Atlas
BenchmarkActivecategory · agenticfamily · terminal-bench · variant 1.0

Terminal-Bench

tbench.ai

terminal tasks solved by agents

data quality57

Updated 3 h ago · first seen 11 Sept 2026

Metric
accuracy · % ↑
Current results
1,639
Models
376
Current leader
gpt-5.6-sol 65.9%

Score history · deepseek-v4-flash-0420 10 rows

Score history for deepseek-v4-flash-04200%20%40%60%80%Sept 26Sept 26Sept 26Sept 26Sept 26Sept 26Sept 26Sept 26Sept 26
  • deepseek-v4-flash-0420
  • 38.64%aa_slug=deepseek-v4-flash-0420-high · variant=hard · evaluator=Artificial Analysis · index_version=4.312 Sept 2026
  • 56.93%aa_slug=deepseek-v4-flash-0420-high · variant=v2.1 · evaluator=Artificial Analysis · index_version=4.312 Sept 2026
  • 3.03%aa_slug=deepseek-v4-flash-0420-high · variant=v4.0 · evaluator=Artificial Analysis · index_version=4.312 Sept 2026
  • 34.09%aa_slug=deepseek-v4-flash-0420-non-reasoning · variant=hard · evaluator=Artificial Analysis · reasoning=off12 Sept 2026
  • 35.61%aa_slug=deepseek-v4-flash-0420 · variant=hard · evaluator=Artificial Analysis · index_version=4.312 Sept 2026
  • 61.8%aa_slug=deepseek-v4-flash-0420 · variant=v2.1 · evaluator=Artificial Analysis · index_version=4.312 Sept 2026
  • 2.53%aa_slug=deepseek-v4-flash-0420 · variant=v4.0 · evaluator=Artificial Analysis · index_version=4.312 Sept 2026
  • 35.61%aa_slug=deepseek-v4-flash-0420 · variant=hard · evaluator=Artificial Analysis · index_version=4.311 Sept 2026
  • 61.8%aa_slug=deepseek-v4-flash-0420 · variant=v2.1 · evaluator=Artificial Analysis · index_version=4.311 Sept 2026
  • 2.53%aa_slug=deepseek-v4-flash-0420 · variant=v4.0 · evaluator=Artificial Analysis · index_version=4.311 Sept 2026

Back to the leaderboard

Frontier over time · accuracy · variant=hard · evaluator=Artificial Analysis

10 leader changes recorded, all dated 11 Sept 2026 — the frontier line needs at least two distinct dates. The corpus is young: every result was first observed on the same day, so leader changes will separate in time as sources are re-crawled.

  1. 65.9%gpt-5.6-sol OpenAI Independent11 Sept 2026
  2. 61.4%gpt-5.6-sol OpenAI Independent11 Sept 2026
  3. 53.0%Claude Sonnet 4.6 Anthropic Independent11 Sept 2026
  4. 51.5%Claude Opus 4.7 Anthropic Independent11 Sept 2026
  5. 50.8%Z.ai GLM 5.2 Z.ai (Zhipu AI) Independent11 Sept 2026
  6. 49.2%KAT-Coder-Pro V2 Kwaipilot Independent11 Sept 2026
  7. 33.3%GPT-5.1-Codex Mini OpenAI Independent11 Sept 2026
  8. 26.5%Grok 4.3 xAI Independent11 Sept 2026

Includes closed rows (history). A point is emitted whenever a result beats every earlier result of the same group, ordered by evaluated_at when the source publishes it, else observed_at.

Leaderboard 317 models · trust independent-evaluator

Select models with +, then Compare.

Leaderboard
#ModelScoreTrustConfigurationvs leaderEvaluatedSourceActions
198devstral-smallOpen weightsMistral AI · Devstral · best of 2 rows6.06%IndependentreasoningoffPartially comparable0.00 ptobs. 12 Sept 2026artificialanalysis.aiT2History
198ernie-4-5-300b-a47bOpen weightsBaidu · ERNIE 4.5 · best of 2 rows6.06%IndependentreasoningoffPartially comparable0.00 ptobs. 12 Sept 2026artificialanalysis.aiT2History
198mistral-large-2Open weightsMistral AI · Mistral · best of 2 rows6.06%IndependentreasoningoffPartially comparable0.00 ptobs. 12 Sept 2026artificialanalysis.aiT2History
198nova-proClosedAmazon Web Services · Nova · best of 2 rows6.06%IndependentreasoningoffPartially comparable0.00 ptobs. 12 Sept 2026artificialanalysis.aiT2History
198qwen3-235b-a22b-instructOpen weightsAlibaba Group · Qwen3 · best of 4 rows6.06%IndependentreasoningonPartially comparable0.00 ptobs. 12 Sept 2026artificialanalysis.aiT2History
198qwen3-30b-a3b-2507Open weightsAlibaba Group · Qwen3 · best of 4 rows6.06%IndependentreasoningoffPartially comparable0.00 ptobs. 12 Sept 2026artificialanalysis.aiT2History
207llama-nemotron-super-49b-v1-5Open weightsNVIDIA · Llama · best of 4 rows5.30%IndependentreasoningonPartially comparable-0.76 ptobs. 12 Sept 2026artificialanalysis.aiT2History
207qwen3-14b-instructOpen weightsAlibaba Group · Qwen3 · best of 3 rows5.30%IndependentreasoningoffPartially comparable-0.76 ptobs. 12 Sept 2026artificialanalysis.aiT2History
207step-3-vl-10bOpen weightsStepFun · Step3 · best of 2 rows5.30%IndependentreasoningonPartially comparable-0.76 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210Hermes-4-70BRestricted weightsNous Research · Hermes 44.55%Independentgroup defaultsPartially comparable-1.51 ptobs. 11 Sept 2026artificialanalysis.aiT2History
210LFM2.5-8B-A1BOpen weightsLiquid AI · LFM2.5 · best of 2 rows4.55%IndependentreasoningonPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210Magistral Small 1.2Open weightsMistral AI · Magistral · best of 2 rows4.55%IndependentreasoningonPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210Ministral 3 14BOpen weightsMistral AI · Ministral 3 · best of 2 rows4.55%IndependentreasoningoffPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210Ministral 3 8BOpen weightsMistral AI · Ministral 3 · best of 2 rows4.55%IndependentreasoningoffPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210Qwen2.5 72B InstructOpen weightsQwen · Qwen2.5 · best of 2 rows4.55%IndependentreasoningoffPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210hermes-4-llama-3-1-70bOpen weightsNous Research · Llama 3.1 · best of 3 rows4.55%IndependentreasoningonPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210llama-3-1-nemotron-instruct-70bOpen weightsNVIDIA · Llama 3.1 · best of 2 rows4.55%IndependentreasoningoffPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210magistral-smallOpen weightsMistral AI · Magistral · best of 2 rows4.55%IndependentreasoningonPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210nvidia-nemotron-nano-12b-v2-vlOpen weightsNVIDIA · Nemotron · best of 4 rows4.55%IndependentreasoningonPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210qwen3-4b-2507-instructOpen weightsAlibaba Group · Qwen3 · best of 4 rows4.55%IndependentreasoningoffPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
210solar-pro-2ClosedUpstage · Solar · best of 4 rows4.55%IndependentreasoningoffPartially comparable-1.51 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222Gemini 2.0 FlashClosedGoogle · Gemini 2.0 · best of 2 rows3.79%IndependentreasoningoffPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222Gemma 3 27BOpen weightsGoogle · Gemma 3 · best of 2 rows3.79%IndependentreasoningoffPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222Mistral Medium 3ClosedMistral AI · Mistral · best of 2 rows3.79%IndependentreasoningoffPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222Phi 4Open weightsMicrosoft · Phi4 · best of 2 rows3.79%IndependentreasoningoffPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222Qwen3 14BOpen weightsQwen · Qwen33.79%Independentgroup defaultsPartially comparable-2.27 ptobs. 11 Sept 2026artificialanalysis.aiT2History
222Qwen3 VL 8B InstructOpen weightsQwen · Qwen3 · best of 4 rows3.79%IndependentreasoningonPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222Qwen3.5-2BOpen weightsQwen · Qwen3.5 · best of 4 rows3.79%IndependentreasoningoffPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222exaone-4-0-32bOpen weightsLG AI Research · EXAONE 4.0 · best of 4 rows3.79%IndependentreasoningonPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222gpt-4.1-nanoClosedOpenAI · GPT 4.1 · best of 2 rows3.79%IndependentreasoningoffPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222motif-2-12-7bClosedMotif Technologies · best of 2 rows3.79%IndependentreasoningonPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
222qwen3-omni-30b-a3b-instructOpen weightsAlibaba Group · Qwen3 · best of 4 rows3.79%IndependentreasoningonPartially comparable-2.27 ptobs. 12 Sept 2026artificialanalysis.aiT2History
233Llama 3.3 70BOpen weightsMeta AI · Llama 3.3 · best of 2 rows3.03%IndependentreasoningoffPartially comparable-3.03 ptobs. 12 Sept 2026artificialanalysis.aiT2History
233Llama-3.1-70BOpen weightsMeta AI · Llama 3.1 · best of 2 rows3.03%IndependentreasoningoffPartially comparable-3.03 ptobs. 12 Sept 2026artificialanalysis.aiT2History
233MiniMax-M1-80kOpen weightsMiniMax · MiniMax · best of 2 rows3.03%IndependentreasoningonPartially comparable-3.03 ptobs. 12 Sept 2026artificialanalysis.aiT2History
233Qwen3 32BOpen weightsQwen · Qwen33.03%Independentgroup defaultsPartially comparable-3.03 ptobs. 11 Sept 2026artificialanalysis.aiT2History
233gemma-4-E2BOpen weightsGoogle · Gemma 4 · best of 4 rows3.03%IndependentreasoningonPartially comparable-3.03 ptobs. 12 Sept 2026artificialanalysis.aiT2History
233midm-250-pro-rsnsftClosedKorea Telecom · best of 2 rows3.03%IndependentreasoningonPartially comparable-3.03 ptobs. 12 Sept 2026artificialanalysis.aiT2History
233qwen3-32b-instructOpen weightsAlibaba Group · Qwen33.03%IndependentreasoningonPartially comparable-3.03 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240Claude Haiku 3.5ClosedAnthropic · Claude · best of 2 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240MiniMax-M1-40kOpen weightsMiniMax · MiniMax · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240Qwen3 8BOpen weightsQwen · Qwen32.27%Independentgroup defaultsPartially comparable-3.79 ptobs. 11 Sept 2026artificialanalysis.aiT2History
240falcon-h1r-7bOpen weightsTII UAE · Falcon · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240gemma-3n-e4bOpen weightsGoogle · Gemma 3 · best of 2 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240granite-4-0-h-smallOpen weightsIBM · Granite 4.0 · best of 2 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240granite-4.1-30bOpen weightsIBM · Granite 4.1 · best of 2 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240granite-4.1-3bOpen weightsIBM · Granite 4.1 · best of 2 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240jamba-1-7-largeOpen weightsAI21 Labs · Jamba 1.7 · best of 2 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240llama-3-1-nemotron-ultra-253b-v1-reasoningOpen weightsNVIDIA · Llama 3.1 · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240mi-dm-k-2-5-pro-dec28ClosedKorea Telecom · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240qwen3-8b-instructOpen weightsAlibaba Group · Qwen3 · best of 3 rows2.27%IndependentreasoningoffPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240sarvam-30bOpen weightsSarvam · best of 2 rows2.27%Independentreasoning_efforthighPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240sarvam-m-reasoningOpen weightsSarvam · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240solar-open-100b-reasoningOpen weightsUpstage · Solar · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
240tri-21b-think-previewOpen weightsTrillion Labs · best of 2 rows2.27%IndependentreasoningonPartially comparable-3.79 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256DeepSeek-R1-0528-Qwen3-8BOpen weightsDeepSeek · Qwen3 · best of 2 rows1.52%IndependentreasoningonPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256Granite 4.0 MicroOpen weightsIBM · Granite 4.0 · best of 2 rows1.52%IndependentreasoningoffPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256Llama 4 ScoutOpen weightsMeta AI · Llama 4 · best of 2 rows1.52%IndependentreasoningoffPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256NVIDIA-Nemotron-Nano-9B-v2Open weightsNVIDIA · Nemotron · best of 4 rows1.52%IndependentreasoningonPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256Qwen3-VL-4B-InstructOpen weightsQwen · Qwen3 · best of 4 rows1.52%IndependentreasoningonPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256R1 Distill Llama 70BOpen weightsDeepSeek · Llama · best of 2 rows1.52%IndependentreasoningonPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256nova-microClosedAmazon Web Services · Nova · best of 2 rows1.52%IndependentreasoningoffPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256olmo-3-32b-thinkOpen weightsAllen Institute for AI · OLMo 3 · best of 2 rows1.52%IndependentreasoningonPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
256sarvam-105bOpen weightsSarvam · best of 2 rows1.52%Independentreasoning_efforthighPartially comparable-4.54 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265Claude 3 HaikuClosedAnthropic · Claude · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265Command AOpen weightsCohere · Command · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265Gemma 3 12BOpen weightsGoogle · Gemma 3 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265Gemma 3 4BRestricted weightsGoogle · Gemma 3 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265Llama 3.1 8BRestricted weightsMeta AI · Llama 3.1 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265Olmo-3-7B-ThinkOpen weightsAllen Institute for AI · OLMo 3 · best of 2 rows0.76%IndependentreasoningonPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265gemma-3n-e2bOpen weightsGoogle · Gemma 3 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265jamba-reasoning-3bOpen weightsAI21 Labs · Jamba · best of 2 rows0.76%IndependentreasoningonPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265lfm2-2-6bOpen weightsLiquid AI · LFM2.2 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265ling-mini-2-0Open weightsinclusionAI · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265llama-3-2-instruct-11b-visionOpen weightsMeta AI · Llama 3.2 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265llama-3-instruct-70bOpen weightsMeta AI · Llama 3 · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265nova-liteClosedAmazon Web Services · Nova · best of 2 rows0.76%IndependentreasoningoffPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
265tri-21b-think-v0-5Open weightsTrillion Labs · best of 2 rows0.76%IndependentreasoningonPartially comparable-5.30 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279LFM2-1.2BOpen weightsLiquid AI · LFM2.1 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279LFM2.5-1.2B-InstructOpen weightsLiquid AI · LFM2.5 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279LFM2.5-VL-1.6BOpen weightsLiquid AI · LFM2.5 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279Llama-3.2-1BRestricted weightsMeta AI · Llama 3.2 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279Ministral 3 3BOpen weightsMistral AI · Ministral 3 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279Molmo2-8BOpen weightsAllen Institute for AI · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279Olmo-3-7B-InstructOpen weightsAllen Institute for AI · OLMo 3 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279Qwen3-0.6BOpen weightsQwen · Qwen3.00%Independentgroup defaultsPartially comparable-6.06 ptobs. 11 Sept 2026artificialanalysis.aiT2History
279Qwen3.5-0.8BOpen weightsQwen · Qwen3.5 · best of 4 rows0%IndependentreasoningonPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279Reka Flash 3Open weightsrekaai · best of 2 rows0%IndependentreasoningonPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279apertus-70b-instructOpen weightsSwiss AI Initiative · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279apertus-8b-instructOpen weightsSwiss AI Initiative · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279exaone-4-0-1-2bOpen weightsLG AI Research · EXAONE 4.0 · best of 4 rows0%IndependentreasoningonPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279gemma-3-1bOpen weightsGoogle · Gemma 3 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279gemma-3-270mOpen weightsGoogle · Gemma 3 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279granite-3-3-8b-instructOpen weightsIBM · Granite 3.3 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279granite-4-0-350mOpen weightsIBM · Granite 4.0 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279granite-4-0-h-350mOpen weightsIBM · Granite 4.0 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279granite-4-0-h-nano-1bOpen weightsIBM · Granite 4.0 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279granite-4-0-nano-1bOpen weightsIBM · Granite 4.0 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279granite-4.1-8bOpen weightsIBM · Granite 4.1 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History
279jamba-1-7-miniOpen weightsAI21 Labs · Jamba 1.7 · best of 2 rows0%IndependentreasoningoffPartially comparable-6.06 ptobs. 12 Sept 2026artificialanalysis.aiT2History

One row per canonical model — its best current row inside this comparability group (effort variants are folded into the model). Bars are relative to the page's best score. “vs leader” reads comparability: partially comparable = same task, conditions differ (reasoning effort, temperature, judge). Rules →