whatmodel.sh / Claude 1.3

Claude 1.3

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Claude 1.3 is a closed-weights language model from Anthropic, released May 2023. Licence: proprietary.

Catalogue identity notes

Anthropic API model claude-v1.3 (later id claude-1.3). Absent from the archived API reference of 2023-03-15, listed in the snapshot of 2023-05-11 ('A significantly improved version of claude-v1'); Anthropic published no launch date, so 'released' is the month it is first documented (true launch between 2023-03-15 and 2023-05-11). Anthropic's Claude 2 model card calls it one of the deployed 'Claude models'. Retired 2024-11-06. Comparison-only.

Model information sources

Release date ↗

48.9±1.3comparison only: no current run option verifiedfrom 15 benchmark results across 7 sources

Score within the LMArena (text) comparison set only.

$ how it compares

The LLMs ranked either side of Claude 1.3, out of 877 with a score, then more from Anthropic.

#modelscorewhereRAMreleased
464Qwen2 57B-A14B Instruct Alibaba49±2local~35GB2024-06
465Granite 4.0 H Tiny (7B-A1B) IBM49±1local~4.5GB2025-09
466Claude 1.3 Anthropic49±1comparison only2023-05
467MiniCPM3-4B OpenBMB49±2local~4GB2024-09
468Bielik 4.5B v3.0 Instruct SpeakLeash49±2local~4.3GB2025-04
more from Anthropic
2Claude Opus 5.5 Anthropic85±1api $8.00/M2026-09
4Claude Fable 5.1 Anthropic83±1api $20.00/M2026-09
6Claude Sonnet 5.5 Anthropic82±1api $4.00/M2026-09

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▅▇█▇▅▄▂▁ #413 of 529 * · 11% of its score · 1 test

Beats LFM2.5 8B-A1B, Qwen3-VL 2B Thinking here, though they rank higher overall. Trails Code Tulu 2 13B, Phi-3-mini-128k-instruct (Apr 2024), DeepSeek LLM 67B Chat here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
HumanEval lab56.0% ✎▁▁▁▁▂▂▆▆█▅▂▁ beats 107/205easy · avg 71%▇▇▇▇ 11%sharp · at its level
math▁▂▃▆█▇▅▇▃▁▁▁ #246 of 523 · 31% of its score · 1 test

Beats MiniMax-M2.1, DeepSeek V3 (Dec '24), GPT-4o (2024-08-06) here, though they rank higher overall. Trails Phi-2 (2.7B), Granite 8B Code Base (4K), Mixtral 8x7B v0.1 (base) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K +281.8% ✎▂▂▂▂▃▄▅▇██▆▄ beats 154/200very easy · avg 90%▇▇▇▇▇▇▇▇▇▇▇▇▇ 31%sharp · several runs
other tests11 not in a task · 57% of its score
testresultvs the others in its rundifficultycountswhy
ARC Challenge lab90.0% ✎▄▁▁▂▁▄█▇▃▂▃▃ beats 136/140very easy · avg 90%▇▇▇▇ 11%sharp · small run
MMLU lab77.0% ✎▁▁▁▁▁▂▂▅██▆▄ beats 158/214easy · avg 80%▇▇▇▇ 10%at its level · small run
BIG-Bench Hard67.3% ✎▁▁▁▁▂▄▅▆█▃▃▁ beats 106/155medium · avg 69%▇▇▇▇ 9.3%sharp · at its level · small run
clembench v1.0 clemscore (self-play)37.1% ✎▄······▄·█·▄ beats 3/4medium · avg 38%▇▇▇ 7.8%sharp · at its level
ZeroSCROLLS Qasper (zero-shot F1)52.3% ✎▂▂··▂·▂··▂█▄ beats 8/10medium · avg 56%▇▇ 4.1%blunt · at its level
TriviaQA lab86.7% ✎▁▁▂▂▄▄▃▆▅█▅▃ beats 79/80easy · avg 84%▇▇ 4.1%small run
ZeroSCROLLS QuALITY (zero-shot accuracy)84.8% ✎▃····▃·▃▃█▅▅ beats 9/10easy · avg 81%▇ 3.6%at its level
SummEdits factual-consistency detection (zero-shot, balanced accuracy)59.8% ✎▄····▄·█···▄ beats 2/4hard · avg 65%▇ 3.6%at its level · noisy
3 more tests
testresultvs the others in its rundifficultycountswhy
ZeroSCROLLS NarrativeQA (zero-shot F1)32.6% ✎▃··▃·█▃·█▃·▃ beats 10/10hard · avg 29%▇ 2.3%blunt · at its level
ConvRe Re2Text, normal relation, zero-shot (prompt 1#)89.7% ✎▂▂▄··▂▂█▄··▂ beats 10/12easy · avg 90%▇ 1.3%noisy
ZeroSCROLLS MuSiQue (zero-shot F1)36.1% ✎▄·█··██·▄▄▄▄ beats 6/10medium · avg 43%▇ 0.6%blunt · at its level

$ where to run it

No verified API or download recorded yet · add one

$ how the score was worked out · Anthropic system cards and mo… 48%, Chain-of-Thought Hub: A Conti… 17%, LLMLingua: Compressing Prompt… 11% and 4 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Anthropic system cards and model announcements (lab-reported scores) · lab-reported5Factual vendor-reported score (not licensed content)
Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance — authors' own runs2CC BY 4.0
LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models — authors' own runs1CC BY 4.0
ZeroSCROLLS: A Zero-Shot Benchmark for Long Text Understanding — authors' own runs4CC BY 4.0
clembench: Using Game Play to Evaluate Chat-Optimized Language Models as Conversational Agents — authors' own runs1CC BY 4.0
SummEdits: Measuring LLM Ability at Factual Reasoning Through The Lens of Summarization — authors' own runs1CC BY 4.0
An Investigation of LLMs' Inefficacy in Understanding Converse Relations — authors' own runs1CC BY 4.0
prior · pull toward the middle
weighted average= 48.9

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json