←Search results

English text is being prepared.

By relevance · Total 1585 results Clear search

Sunday

  7. Story4 reports · 2 facts

    Fireworks evaluates DeepSeek's new model

    On September 14, 2026, Fireworks AI released and evaluated DeepSeek-V4.1-Flash. According to its published results, the model achieves 74.34% pass@1 on DeepSWE, on par with GPT-6-Astra's 74.12%; its cost per task is $0.43, about 1/15 that of Astra. The model is a 552B-parameter MoE, using separate activation budgets of 8B for input and 16B for output, and its KV Cache HBM footprint is reduced to 1/4 that of the previous-generation V4. On HLE it scores 34.52%, below Astra's 50.40%, and Fireworks considers it better suited to agentic coding than to highly difficult reasoning. On September 28, Fireworks AI released its first cache-aware router, FireRouter with Opus, which is now live in the CLI and offered for the first time as a standalone routing model; it can select among Claude Opus 5.5, GLM 5.3 and GLM 5.3 Flash based on task fit, prompt cache cost and switching cost; its internal A/B testing over more than a month showed coding task cost per session dropping from $15.36 to $6.63, a 57% reduction (±19 percentage points), with accuracy at pure O

    On September 14, 2026, Fireworks AI released and evaluated DeepSeek-V4.1-Flash. According to its published results, the model achieves 74.34% pass@1 on DeepSWE, on par with GPT-6-Astra's 74.12%; its cost per task is $0.43, about 1/15 that of Astra. The model is a 552B-parameter MoE, using separate activation budgets of 8B for input and 16B for output, and its KV Cache HBM footprint is reduced to 1/4 that of the previous-generation V4. On HLE it scores 34.52%, below Astra's 50.40%, and Fireworks

    Matched fieldsSummary

  9. Story2 reports · 1 facts

    Fireworks AI releases Ember-1 based on Kimi K3

    On September 23, 2026, Fireworks AI released Ember-1, an efficient inference model based on Kimi K3. The company says the model cuts token consumption by about 40% while maintaining equivalent quality; across seven benchmarks and production traffic from two customers, K3's reasoning process can be shortened by 35–50% without any drop in accuracy. Ember-1 is available on Serverless as a Research Preview, offered alongside the base Kimi K3, with training support on enterprises' own data also opened at the same time. Follow-up reports say the model is offered only through the Fireworks serverless API, with no public weights, training code, or specific training algorithm, and no self-hosting support for now; in A/B tests against two customers' production coding workloads, output tokens per task fell from 49.3K to 29.9K, reasoning tokens dropped 71.3%, total tokens fell 39%, and task scores were essentially flat at 0.753 versus 0.751, with one customer already using it in production. The reports also say Ember-1's per-token price is the same as Kimi K3's, with the savings coming from generating fewer tokens.

    On September 23, 2026, Fireworks AI released Ember-1, an efficient inference model based on Kimi K3. The company says the model cuts token consumption by about 40% while maintaining equivalent quality; across seven benchmarks and production traffic from two customers, K3's reasoning process can be shortened by 35–50% without any drop in accuracy. Ember-1 is available on Serverless as a Research Preview, offered alongside the base Kimi K3, with training support on enterprises' own data also opene

    Matched fieldsSummary

Tuesday

Wednesday

Wednesday

Friday

  1. Story2 reports · 1 facts

    Fireworks AI analyzes DeepSWE v1.1 model routing potential

    Fireworks AI released an analysis on September 21, 2026, testing 18 models on 113 coding tasks in DeepSWE v1.1 to examine the potential of routing models by task. Its published results show that selecting the best model per task after the fact can reach a 97.6% pass rate at a cost of $1.88 per task, while the best-performing single model, GPT-6 Astra, has a pass rate of 74.1% at a cost of $6.52 per task. If limited to the open-weight model pool, routing can still reach 90.3% at $1.45 per task. The authors also note that these figures come from oracle routing selected after the fact, which represents an upper-bound perspective; a real router must make choices before seeing the results. Its FireRouter, across 2,334 production coding sessions over four weeks, reduced cost from $15.81 for the single model Opus 5 to $7.42.

    Fireworks AI released an analysis on September 21, 2026, testing 18 models on 113 coding tasks in DeepSWE v1.1 to examine the potential of routing models by task. Its published results show that selecting the best model per task after the fact can reach a 97.6% pass rate at a cost of $1.88 per task, while the best-performing single model, GPT-6 Astra, has a pass rate of 74.1% at a cost of $6.52 per task. If limited to the open-weight model pool, routing can still reach 90.3% at $1.45 per task. T

    Matched fieldsSummary

Monday