Jamba 1.5 Mini
Aug 2024
Jamba 1.5 MiniCompare inon
Benchmarks of Jamba 1.5 Mini14
BenchmarkByScoreRank
Knowledge13
Health & law1
No benchmark matches
—per task?
price not verifiedprice
—output speed?
256Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-5 Minihighwins 71%—$0.053
OpenGemma 4 31Bunstatedwins 79%—$0.10 / $0.34BetterGPT-6 Lunamaxwins 73%—$0.068Function calling only with none · Chat Completions+3
BetterClaude Haiku 5.5highwins 74%—$0.079Remove sampling parameters+5BetterGPT-5.6 Lunaxhighwins 78%—$0.085ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-5 Minihighwins 71%—$0.053GPT-6 Lunamaxwins 73%—$0.068Function calling only with none · Chat Completions+3Claude Haiku 5.5highwins 74%—$0.079Remove sampling parameters+5GPT-5.6 Lunaxhighwins 78%—$0.085GPT-5.6 Terraxhighwins 78%—$0.63Prompts with >272K input tokensModelScoreCostPer taskWhat changes
Gemma 4 31BunstatedOpenwins 79%—$0.10 / $0.34
Qwen3.8 27BunstatedOpenwins 78%—$0.42 / $2.55max_tokens range · QwenCloud+4Gemma 4 26B A4BunstatedOpenwins 78%—$0.15 / $0.60
Mistral Medium 3.5unstatedOpenwins 77%—$1.50 / $7.50Memory (4-bit): 31 GB → 77 GBQwen 3.5 122B A10BunstatedOpenwins 76%—$0.40 / $3.20Memory (4-bit): 31 GB → 75 GBRun it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✓ 48 GB GPU · compressed✓ 80 GB GPU · 8-bit
Licencejamba-open-model-licensecommercial use with conditions
Benchmarks
all categoriesChat & Knowledge
Arena Text · overall1,240#177
Arena Text · Chinese1,135.7#184
Arena Text · English1,223.9#181
Arena Text · German1,162#151
Arena Text · instruction following1,147.9#189
Arena Text · Japanese1,019.2#144
Arena Text · non-English1,134.7#189
Arena Text · Russian1,156.2#189
Vals LegalBench66.6#86
Arena Text · overall · October 22, 20261,186.6#179
Arena Text · overall · October 15, 20261,186.6#170
Writing & Language
Arena Text · creative writing1,211#170
FAQ
How much does Jamba 1.5 Mini cost?›
There is no per-unit API price: price not verified.
How good is Jamba 1.5 Mini, and what is it best at?›
#181 of 205 in Chat & Knowledge, #178 of 197 in Writing & Language.
Is Jamba 1.5 Mini better than Phi 4?›
On Arena Text · English, Jamba 1.5 Mini scores 1,223.9 vs 1,230.9 — about the same.
What is the context window of Jamba 1.5 Mini?›
256K tokens.
When was Jamba 1.5 Mini released? Is there a newer version?›
Released 2024-08-22. It is the current version in its line.
Is Jamba 1.5 Mini open source?›
Yes, the weights are open (jamba-open-model-license licence).
Can I run Jamba 1.5 Mini locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Jamba 1.5 Mini read images?›
No, it accepts text only.
What is Jamba 1.5 Mini?›
Jamba 1.5 Mini is an open-weights language model from AI21 Labs, released Aug 2024.
What are the best alternatives to Jamba 1.5 Mini?›
GPT-5 Mini (better), Gemma 4 31B (open weights).